Morgan v. Family Dollar Stores, Inc.

HULL, Circuit Judge: The Court sua sponte issues this corrected opinion. An opt-in class of 1,424 store managers, in a collective action certified by the district court, sued Family Dollar Stores, Inc. (“Family Dollar”) for unpaid overtime wages under the Fair Labor Standards Act (“FLSA”), 29 U.S.C. §§ 201-219. During an eight-day trial, the Plaintiffs used Family Dollar’s payroll records to establish that 1,424 store managers rou *1240 tinely worked 60 to 70 hours a week and to quantify the overtime wages owed to each Plaintiff. Family Dollar focused on its affirmative defense that the store managers were executives within the meaning of the FLSA and exempt from its overtime pay requirements. The jury found that the Plaintiff store managers were not exempt executives and that Family Dollar had willfully denied them overtime pay. The jury awarded $19,092,003.39 in overtime wages. The court entered a final judgment of $35,576,059.48 against Family Dollar consisting of $17,788,029.74 in overtime wages and an equal amount in liquidated damages. Because of the complex procedural history from 2001 to 2005 that led to the case being certified as a collective action, the subsequent eight-day trial in 2006, and Family Dollar’s myriad challenges on appeal, we preface the opinion with a table of contents: I.PROCEDURAL HISTORY FROM 2001-2005 .1241 A. Complaint.1241 B. April 2001 Motion to Facilitate Nationwide Notice.1242 C. October 2001 — Second Motion to Facilitate Nationwide Notice.1242 D. July 2002 Notice.1242 E. October 2002 — Third Motion to Facilitate Nationwide Notice.1242 F. November 2002 Order and Fact Findings.1243 G. December 2002 Notice to Potential OpVIns.1243 H. Discovery Disputes.1243 I. May 2004 Motion to Decertify the Collective Action.1245 J. January 2005 Order and Fact Findings.1246 K. First Jury Trial.1247 II.SECOND JURY TRIAL IN 2006.1247 A. Corporate Structure.1248 B. Store Managers.1248 C. Family Dollar Executives .1251 D. District Managers.1254 E. Salary Compared to Hourly Wages.1257 F. Judgment/Verdict.1258 III. DECERTIFICATION.1258 A. FLSA’s Similarly Situated Requirement.1258 B. Two-Stage Procedure for Determining Certification.1260 C. District Court’s Denial of Decertification.1262 IV.EXECUTIVE EXEMPTION DEFENSE.1265 A. FLSA’s Executive Exemption.1265 B. Primary Duty Is Management.1266 C. Family Dollar’s Motion for Judgment as a Matter of Law.1269 D. Other Circuits’ Cases.1271 E. 163 Individual Plaintiffs Granted Judgment on Executive Exemption Defense.1273 V. REPRESENTATIVE TESTIMONY .1276 VI. WILLFULNESS AND LIQUIDATED DAMAGES .1280 A. Willful Violation.1280 B. Good Faith and Liquidated Damages .1282 VII. JURY INSTRUCTIONS.1283 VIII. CONCLUSION.1285 *1241 I. PROCEDURAL HISTORY FROM 2001-2005 A. Complaint Family Dollar is a nationwide retailer that operates over 6,000 discount stores that sell a wide assortment of products, including groceries, clothing, household items, automotive supplies, general merchandise, and seasonal goods. 1 In January 2001, Janice Morgan and Barbara Richardson, two store managers, filed a Complaint on behalf of themselves “and all other similarly situated persons,” alleging that Family Dollar willfully violated the FLSA by refusing to pay its store managers overtime compensation. The first Complaint asserted that Family Dollar paid store managers a salary, required them to work 60 to 90 hours a week, and refused to compensate them for overtime. According to Plaintiffs, store managers are managers only in name and actually spend the vast majority of their time performing manual labor, such as stocking shelves, running the cash registers, unloading trucks, and cleaning the parking lots, floors, and bathrooms. Store managers spend only five to 10 hours of their time managing anything. Plaintiffs sought unpaid benefits, overtime compensation, and liquidated damages due to Family Dollar’s willful FLSA violations. The Complaint urged the district court to issue notice of the action to all similarly situated Family Dollar employees nationwide, and to inform them of their right to opt into the suit as a collective action. Plaintiffs relied on 29 U.S.C. § 216(b), which authorizes courts to maintain a case as one collective action so long as the employee-plaintiffs are similarly situated. Family Dollar’s Answer raised a number of affirmative defenses. It asserted that its store managers were exempt executives 2 and denied any violations were willful. Family Dollar also argued that a collective action, under § 216(b), was impermissible because (1) the store managers were not similarly situated, (2) Plaintiffs’ claims were not representative of others in the group, and (3) Plaintiffs could not satisfy § 216(b)’s requirements for maintaining a collective action. In May 2001, Plaintiffs filed their Third Amended Complaint on behalf of Morgan and Richardson, and added Cora Cannon and Laurie Trout-Wilson as Plaintiffs. The Third Amended Complaint raised the same claims for overtime pay and, again, urged the district court to notify other similarly situated store managers of the action. B. April 2001 Motion to Facilitate Nationwide Notice In April 2001, Plaintiffs moved the district court to (1) certify the case as a collective action, (2) authorize notice “by first class mail to all similarly situated management employees employed by Family Dollar Stores, Inc. at any time during the three years prior to the filing of this action to inform them of the nature of the action and their right to opt-into this lawsuit,” and (3) order Family Dollar to “produce a computer-readable data file containing the names, addresses, Social Security number and telephone numbers of such potential opt-ins so that notice may *1242 be implemented.” In May 2001, the court denied the motion for immediate notice, but indicated the motion was “overruled without prejudice.” In September 2001, the district court issued a scheduling order pursuant to Rule 16(b). 3 The order indicated that the parties mutually agreed to “an initial period of discovery limited to identification of claims and their factual basis,” and that, despite Family Dollar’s opposition, Plaintiffs would request the court to facilitate notice on or before February 2002. Discovery was to expire on October 1, 2002. C. October 2001 — Second Motion to Facilitate Nationwide Notice In October 2001, Plaintiffs renewed their motion to facilitate notice. Family Dollar twice opposed Plaintiffs’ motion and urged the district court to delay ruling to allow more discovery. At oral argument in April 2002, the court withheld ruling pending additional discovery by Family Dollar, and ordered Plaintiffs’ counsel to make all named Plaintiffs available for deposition. In April 2002, before the court ruled on the renewed motion, the parties jointly agreed to send limited notice of the suit to current and former store managers that worked in the regions where the named Plaintiffs worked from July 1, 1999 to the present. As a result, the court denied Plaintiffs’ October 2001 motion as moot. D. July 2002 Notice In July 2002, the parties notified 784 potential class members in Region 4 (which contains 15 Family Dollar districts in Alabama, Mississippi, Louisiana, Georgia, and Tennessee), district 39 (in Georgia), and district 118 (in New York). The jointly-sent notices required the recipients to mail their consent forms by October 22, 2002. In August 2002, the court extended the discovery deadline by 120 days. By October 2002, 142 store managers from different states had filed consent forms. Plaintiffs’ counsel subsequently sent each of those store managers an 11-page questionnaire with 17 questions and 75 total subparts. The questionnaire asked about employment dates, weekly work hours, day-to-day duties, amount of hours spent on manual labor, what independent authority store managers had, whether district managers made all important managerial decisions, whether hourly assistant managers performed the same duties, and a host of other questions relating to Family Dollar store operations. E.October 2002 — Third Motion to Facilitate Nationwide Notice In October 2002, Plaintiffs renewed their motion to facilitate nationwide notice. Plaintiffs’ motion estimated that approximately 11,164 current or former store managers had no notice of the action and that, based on the approximately 20% response rate, a sizeable number of potential class members would opt into the suit. Plaintiffs’ motion and counsel’s affidavit summarized the responses to the 11-page questionnaire. Plaintiffs argued that the opt-ins’ responses showed that the store managers were similarly situated and that there were enough initial responses to warrant nationwide notice. Plaintiffs also offered the Rule 30(b)(6) deposition testimony of Bruce Barkus, the Executive Vice President of Store Operations at Family Dollar. Barkus admitted that the store manager job description is the “current and only job description[ ] for Family Dollar Store Managers,” and acknowledged that Family Dollar made no *1243 inquiry into how many hours a week store managers actually work or whether store managers’ actual day-to-day duties mirror the ones outlined in the job description. Plaintiffs argued that Barkus’s testimony bolstered the questionnaire responses that showed store managers spent 90% of their time on manual tasks. F.November 2002 Order and Fact Findings In November 2002, the district court, acting pursuant to § 216(b), granted Plaintiffs’ motion to facilitate nationwide class notice to “all former and current Store Managers who work and/or worked for the Defendant over the last three years.” The court found Family Dollar’s store managers were similarly situated within the meaning of § 216(b) because they: (1) worked 60 to 80 hours a week; (2) received a fixed salary and no overtime pay; (3) spent 75 to 90% of their time on non-managerial tasks such as stocking shelves, running the cash registers, unloading trucks, and performing janitorial duties; (4) did not consistently supervise two or more employees; (5) lacked the authority to hire, discipline, or terminate employees without first obtaining permission from their district managers; (6) could not select outside vendors without their district managers’ permission; (7) worked no less than 48 hours a week under the threat of pay cuts or loss of leave time; and (8) arrived at work before the store opened and stayed until after closing. Although the district court acknowledged that there existed “some differences between the named-Plaintiffs and the opt-ins in terms of pay scale and job duties,” it concluded that “these differences do not preclude the facilitation of nationwide service.” The court stressed that Plaintiffs must only be “similarly situated” — not “identically situated.” The court considered Family Dollar’s contention that its stores have “different locations, are of various sizes, and sell different volumes of merchandise.” But the court found that those differences did not undermine the factual basis for concluding that Family Dollar’s store managers were similarly situated. The court emphasized that it had the benefit of making its decision after twenty months of litigation, considering Plaintiffs’ motion to facilitate nationwide notice on two previous occasions, and giving Family Dollar an opportunity to depose the named Plaintiffs. The court found that a sufficient number of similarly situated employees likely were interested in joining the suit and that the case could be managed and resolved in a single litigation. G. December 2002 Notice to Potential Opt-Ins In December 2002, Plaintiffs’ counsel mailed 12,145 notices nationwide to current and former Family Dollar store managers employed on or after July 1, 1999. Each had until February 25, 2003 to return the enclosed consent forms. Each mailing included the 11-page questionnaire. By March 2003, nearly 2,500 current and former Family Dollar store managers had joined the litigation. H. Discovery Disputes Throughout the litigation, the district court resolved scores of discovery-related motions. For example, Plaintiffs refused to turn over certain questionnaire responses based on attorney-client privilege. This triggered various motions to compel. In another instance, Family Dollar refused to provide Plaintiffs with information related to the identity of its district managers. Plaintiffs responded with their own motions to compel. Meanwhile, a fight over depositions was brewing. In March 2003, Family Dollar notified Plaintiffs of its in *1244 tent to depose, either in person or by-written questions, all opt-in Plaintiffs. The court extended discovery until August 29, 2003. And in June 2003, in a comprehensive order, the court (1) required Plaintiffs’ counsel to produce the questionnaire responses used to support Plaintiffs’ motion to facilitate nationwide notice; (2) ordered Family Dollar to produce the names, addresses, and telephone numbers of all former Family Dollar district managers since June 1999; (3) prohibited Plaintiffs’ counsel from engaging in ex parte communication with former Family Dollar district managers; and (4) clarified that “this Court shall ... treat each opt-in Plaintiff as a separate party for purposes of enforcement of the Scheduling Order.” Discovery issues continued to surface. The court again extended deadlines for discovery to December 12, 2003, and for dispositive motions to January 12, 2004. In October 2003, Family Dollar informed Plaintiffs that it intended to depose all remaining opt-in Plaintiffs, using written questions, pursuant to Rule 31. In November 2003, Family Dollar’s counsel sent a letter stating it planned to take 2,100 depositions in seven days, using 338 written questions per deponent, from December 6 to 12, 2003, in Birmingham. Plaintiffs moved for protective orders to limit the number of depositions to two a day until the end of discovery and to prevent Family Dollar from deposing the nearly 2,100 opt-in Plaintiffs. Plaintiffs’ motion noted that discovery had been ongoing for two and a half years, and that prior to October 2003, Family Dollar failed to depose any of the opt-in Plaintiffs. Plaintiffs objected to Family Dollar’s attempt to depose 2,100 opt-in Plaintiffs during the last 29 days of discovery as burdensome, unreasonable, and expensive. In November 2003, Family Dollar moved for a protective order, prohibiting disclosure of the written deposition questions to the opt-in Plaintiffs. In January 2004, the district court issued a discovery management order resolving many issues. The court limited Family Dollar to “not more than” 250 depositions of the opt-in Plaintiffs, “including those who [had] already been deposed.” The court did not restrict Family Dollar to written questions, but limited depositions to five per day (each three hours long). The order authorized Plaintiffs to select 250 opt-ins for Family Dollar to depose in-person. The court pushed the discovery deadline back to April 12, 2004, with dis-positive motions due May 12, 2004. It denied both parties’ motions for protective orders as moot. In late January 2004, Family Dollar moved to clarify or alter the discovery management order. In early February 2004, the district court denied Family Dollar’s request to depose the rest of the opt-in Plaintiffs under Rule 31 (written deposition questions), but granted Family Dollar leave to use Rule 33 (interrogatories) to obtain discovery from the remaining opt-in Plaintiffs. In March 2004, the court issued an order clarifying that Family Dollar was entitled to serve 25 interrogatories on every opt-in Plaintiff. By mid-February 2004, 152 opt-in Plaintiffs (of the 250) had not been deposed in-person. The court gave Plaintiffs’ counsel seven days to provide Family Dollar with a list of the remaining 152 opt-in Plaintiffs and the dates that each would be available for deposition in Birmingham. To ensure Family Dollar had an opportunity to depose the remaining 152 opt-in Plaintiffs in Birmingham, the district court, in late February 2004, also ordered Plaintiffs’ counsel to provide Family Dollar with three days notice of any change in scheduled depositions, and threatened to dismiss opt-in Plaintiffs who failed to attend. For the next several months, the court *1245 dismissed with prejudice various opt-in Plaintiffs for their failure to appear at depositions in Birmingham. By the end of discovery, Family Dollar deposed, in person, 250 opt-in Plaintiffs and the named Plaintiffs. 4 In addition to the 250 depositions of the opt-in Plaintiffs, the parties deposed Family Dollar’s executives, district managers, various experts, and other witnesses. Family Dollar produced voluminous payroll records, store manuals, emails, and other communications. Plaintiffs produced the individual responses to the questionnaire. The record was fully developed before the next critical step in this case. I. May 200k Motion to Decertify the Collective Action In May 2004, Family Dollar moved to decertify the collective action under § 216(b). Relying on affidavits and a wealth of information revealed during discovery, the parties briefed whether the case should proceed as 1,424 individual actions 5 or as a § 216(b) collective action. Family Dollar argued (1) the opt-in Plaintiffs were not similarly situated under the FLSA, because their day-to-day job duties were too different; (2) the executive exemption defense is inherently individualized; and (3) discrepancies in the store managers’ duties made a collective trial impossible and unfair. In response, Plaintiffs pointed out that discovery established that all store managers were similarly situated because they (1) have the same job description, (2) spend 75 to 90% of their time on the same non-management duties, and (3) spend little time on the management duties in their job description. In addition to Barkus’s testimony, Plaintiffs emphasized the deposition testimony of two other Family Dollar executives, Bill Broome and Dennis Heskett, indicating that Family Dollar applied the overtime exemption across the board without any consideration of store-by-store variables, and that store size and location did not affect Family Dollar’s decision to exempt all store managers from overtime pay requirements. 6 *1246 J. January 2005 Order and Fact Findings In a January 2005 order, the district court denied Family Dollar’s motion to decertify, determined that none of the factual findings in its November 2002 order had been called into question, and made additional fact-findings. The court’s order also expressly incorporated those 2002 findings by reference. In addition, the court found that the “evidence eonfirm[ed] that substantial similarities exist in the job duties of the named and opt-in Plaintiffs.” The court found that 90% of the named and opt-in Plaintiffs (1) interview and train employees, (2) direct work of employees, and (3) maintain production and sales records. The court also found that the named and opt-in Plaintiffs had similar restrictions on the scope of their responsibilities. Although classified as store managers, they lacked independent authority to hire, promote, discipline, or terminate assistant managers; award employees pay raises; or change weekly schedules of hourly employees. And 90% lacked the power to close the store in an emergency without the district manager’s permission. The court concluded that none of the named and opt-in Plaintiffs were responsible for the “total operation of their stores,” and that, in reality, district managers performed the relevant managerial duties. The court found: [m]uch of the management discretion which would ordinarily be exercised by store managers is exercised by the district management pursuant to corporate policies and practices set at headquarters. The store managers are very closely supervised by the district managers. In terms of managerial duties, the district manager is more directly responsible for the operation of a Family Dollar store than the store manager. The court also determined that Plaintiffs similarly spent their time between managerial and non-managerial duties. It found that most (90%) of the named and opt-in Plaintiffs (1) “spend only a small fraction of their time performing managerial duties”; (2) “spend the vast majority of their time on essentially non-managerial duties such as unloading trucks, stocking shelves, working as cashiers, and performing janitorial duties”; and (3) shared some of their managerial duties with nonexempt, hourly employees (ordering merchandise, controlling store keys, opening and closing the store, depositing money in banks, and approving checks, refunds, and returns). Viewing the evidence “as a whole,” the court found that (1) the primary duties of Plaintiffs were not managerial; (2) the time spent performing non-managerial duties did not significantly differ from store to store, district to district, or region to region; and (3) the relative importance of the non-managerial duties (as compared to the limited number of managerial duties) did not vary significantly depending on the store or district. Further, “the basic pay rates of the named and opt-in Plaintiffs are also similar,” in that Family Dollar paid “all of its store managers a base salary regardless of the weekly hours they work.” The court determined that *1247 Family Dollar’s defenses were not so individually tailored to each Plaintiff that a collective action would be unmanageable. Because substantial similarities existed in the Plaintiffs’ job duties, and the same policies and procedures applied to each store, the court concluded that the case could be fairly tried as a collective action. 7 K. First Jury Trial In February 2005, the court entered a pretrial order summarizing each party’s trial position. Family Dollar did not contest that it engaged in interstate commerce or that it failed to pay its store managers time and a half for any work over 40 hours a week. Its success at trial would turn on whether Family Dollar proved its store managers were exempt executives. Before the first jury trial, the court allowed the Plaintiffs to represent the opt-in members as a whole. Family Dollar argued that Plaintiffs were not similarly situated, repeatedly sought decertification, and alternatively argued the Plaintiffs should be divided into nine subgroups. The court denied Family Dollar’s requests and allowed the case to be tried as a collective action with the Plaintiffs representing (and thus binding for good or bad) the opt-in members as a whole. In the first trial, the jury deadlocked. Therefore, the issues on appeal arise out of the second trial. II. SECOND JURY TRIAL IN 2006 The second jury trial lasted eight days. This time the jury reached a verdict, expressly finding the store managers were not exempt. The parties called 39 witnesses-store managers, district managers, corporate executives, payroll officials, and expert witnesses. In total, the testifying store managers worked at 50 different Family Dollar stores. The testifying district managers ran the operations of 134 different stores. Two testifying Family Dollar executives oversaw 1,400 stores, while a third testifying executive was in charge of all stores. The parties presented hundreds of Family Dollar’s records detailing its policies and procedures. These records included Family Dollar’s Store Policy Manual, subsequent manual revisions, four volumes of the Professional Development Training Reference Book, the Personnel Training Manual, various Frequently Asked Question documents, “Weekly Work Schedules,” and emails by district managers to store managers. The parties also introduced a large volume of payroll records showing (1) the number of hours worked by each Plaintiff store manager each week, (2) each store manager’s salary and rate of pay, and (3) the number of hours every employee worked each week. Both parties submitted multiple exhibits summarizing payroll data in easy-to-digest charts. Given the jury’s verdict and our standard of review, we outline the trial evidence in the light most favorable to the Plaintiffs. 8 *1248 A. Corporate Structure Family Dollar is a publicly held, nationwide retailer that operates over 6,000 discount stores in 40 states and the District of Columbia. It has annual sales of around $5 billion and annual net profits ranging from $200 to 263 million from 1999 to 2005. Its individual stores have average annual sales of $1 million, and average net profits of 5 to 7%, or $50,000 to $70,000. Family Dollar structures store operations into five divisions (each headed by a vice-president), 22 regions (each headed by a regional vice-president), and 380 districts (each overseen by a district manager). Each district contains multiple stores. Each district manager supervises the operations of 10 to 30 stores. Some districts are small with multiple stores in an urban area. Other districts are larger with small stores in small towns. The district manager’s office is housed within one store in the district. Family Dollar’s corporate office issues instruction manuals with operating policies that apply uniformly to all stores nationwide. No matter the size of the store or the district, every detail of how the store is run is fixed and mandated through Family Dollar’s comprehensive manuals. B. Store Managers Family Dollar has the same job description for all store managers and lists their “Essential Job Functions” as: 1.Supervise all store personnel, including assigning tasks, ensuring compliance with merchandising and operational policies, and locking and unlocking store. 2. Prepare, complete and transmit store reports as required. 3. Count money/checks, prepare bank deposits and travel to bank. 4. Count petty cash, get change from bank, unlock petty cash drawer and give change to cashiers as needed in registers. 5. Post net sales in Beat Yesterday Book. 6. Train Cashiers and Stock Clerks through verbal instructions and non-verbal demonstration. 7. Count stock, calculate amount to order, use MSI machine to order and transmit, calculate additional goods needed for ad bulletins and endcap programs. 8. Read, plan and stock schematics for proper merchandising. 9. Practice cash control policies; including check approvals, refund and exchange approvals, layaway approvals (when applicable). 10. Work as Cashier when needed and be able to perform all Cashier tasks, including but not limited to: • Ring up sales on cash register, receive money/check and count correct change to customer; hold and use register key while on shift. • Scan merchandise labels and place merchandise in bags. *1249 • Count money in cash register when coming on and leaving shift; write and sign the Balance & Declare Slip at end of shift. • Clean cash register area using cleaning solutions and paper towels; sweep mat, dressing room and entry areas as necessary. • Verbally greet each customer as they enter the store; receive bags and staple labels to each bag customer brings into store. • Straighten and restock gum, candy and other areas on register. • Attach labels and software to merchandise using Meto gun; place price stickers on hardliners merchandise; read pre-printed labels and use Swif-tach machine to attach softline merchandise. • Place merchandise on correct clothes hangers, button/zip and hang on rack as needed. • Cut register tapes and tape to appropriate packages. • Unpack shipping packages (totes) and prepare for shelf placement. • Monitor dressing room, including opening door, counting merchandise and checking after customer leaves. • Watch suspected shoplifters; ring front bell to notify other employees of situation needing help. 11. Work as Stock Clerk when needed and be able to perform all Stock Clerk tasks including but not limited to: • Unload incoming freight off trucks; count incoming boxes and verify correct receipt and log in freight log. • Open boxes, check for damaged goods and remove merchandise. • Stock shelves with merchandise, including bulky and heavy goods. • Pack merchandise, complete transfer labels and seal cartons for outgoing shipment. • Organize and maintain layaway bins. • Maintain organization of sign and pricing area. • Store and remove seasonal repacks as necessary. • Perform simple maintenance such as changing light bulbs, moving shelving and panels. The overwhelming evidence showed that Plaintiff store managers exercise little discretion and spend 80 to 90% of their time performing manual labor tasks, such as stocking shelves, running the cash registers, unloading trucks, and cleaning the parking lots, floors, and bathrooms. Even as to the assigned management tasks, such as paperwork, bank deposits, and petty cash, the store manual strictly prescribes them. And district managers closely scrutinize store managers to ensure compliance with the manual and corporate directives. Family Dollar forbids outside janitorial help, and store managers lack authority to hire outside vendors. Store managers, just as hourly employees, are expected to clean the store. For the purposes of “End of Day Recovery,” the manual requires that “[t]wo hours before closing, all employees are to stop their current projects and begin a systematic cleaning and straightening up of the store. (The only exception would be the Cashier who is ringing up sales.) If the store has been shopped heavily, more time may be required to satisfactorily recover.” The manual specifies the trash must be emptied (after checking for hot cigarette butts), the floors must be swept every day, and the floors must be mopped with clear water at least once a week. Rest rooms must be cleaned and mopped daily, stocked with toilet tissue, paper towels, and a trash container that is to be emptied daily. “Under no circumstances should *1250 there be merchandise, equipment or fixtures in the rest rooms.” Store managers routinely perform janitorial duties. The manual even prescribes how janitorial tasks are to be performed: Each morning the sidewalk is to be swept, the parking lot is to be free of debris, the rides wiped clean and checked for safety reasons, the drink machine wiped clean, the trash cans emptied, the window ledges cleaned of cobwebs and dust, and the windows washed as necessary. Weeds must be removed. Do not allow weeds to grow around the building in the front or back. Pay special attention to the sidewalk area in front of the store. Remove any weeds that may be growing here. The back of the building (the entire perimeter, if the building is free standing) is to be maintained just as the front. No shopping carts, fixtures or blue totes are to be stored outside. If there are problems with people using the store’s dumpster or leaving large items behind the store, contact the District Manager for instructions. Sweep mats as often as necessary during the day to maintain a clean and neat appearance. If the mat has spots, they are to be removed with a spot remover. Remove chewing gum by hardening the gum with ice, then pulling the gum off. If the entrance mats get wet, hang them for a few minutes. The water will drain off the mat. With respect to indoor janitorial duties: The windows are to be kept free of cobwebs and dusted daily. The window ledges are also to be maintained daily as needed, free of dust and debris. Store managers lack discretion over the store’s merchandise selection, prices, sales promotions, and layouts — all are set by the home office and district managers. For example, each store is provided a schematic layout and diagram of the store which shows (1) where each shelf must be, (2) what product goes on each shelf, (3) how all merchandise is to be displayed, (4) how all signs, merchandising, and display information is to be used, (5) how each “end cap” (the end of an aisle or gondola) should be displayed, and (6) what promotional product goes on the end cap. Every month, corporate headquarters mails each store a promotional programs booklet that contains the monthly planning calendar and a number of merchandise programs. The manual admonishes that “any deviation from the company program must have the District Manager’s approval.” 9 The tiniest of details are governed by the manuals. For example, the manual’s “Clip Boards in the Office” page details how a store must structure its clip boards. 10 Even the four drawer cabinets, located in every store, are organized identically. 11 The manual also has a subsection on “Break Area and Coffee Pots.” It states that “Appliances such as coffee pots, *1251 microwave ovens, refrigerators, etc., must be approved by the District Manager. Whoever makes the coffee is responsible for unplugging the coffee pot when not in use.” The manual instructs: “Do not use the on/off switch as this can be left on by mistake and create a fire hazard. Make periodic checks throughout the day and before closing to assure that it has been unplugged.” Store managers must follow strict rules regarding store keys, bank deposits, petty cash, and store operating hours. For example, the manual requires there be $300 in petty cash, divided into $200 in the petty cash-box for making register change, and $50 in beginning funds for each of the two registers. In some cases, the petty cash amount may be increased due to the volume of business with the district manager’s and regional vice-president’s approval and notification to the Cash and Sales Department. The manual indicates that “[t]he amount of the store’s Cash Accumulation is to be set by the District Manager .... [and] should be posted in the petty cash box using the ‘Cash Accumulation Card.’ ” Store managers have the same paperwork to do and time frame in which to do it. Further, each store has a preassigned “truck day” when the company truck delivers merchandise to the store. Because of the volume of unloading and stocking, the store manager always works “truck day.” The store manager helps unload 800 to 1,500 cartons from the truck to the storeroom and stock the shelves. The evidence also showed that store managers are assigned a fixed payroll budget, with total labor hours to come from that budget each week, and are required to use only hourly employees. As detailed later, store managers have scant discretion to act independently of their district managers. C. Family Dollar Executives Plaintiffs called two Family Dollar executives who testified about store managers’ roles in the Family Dollar- corporate hierarchy. Bruce Barkus started with Family Dollar in 1999, oversaw all stores, and reported to the President. He testified that Family Dollar classified store managers as executives, across the board, without ever determining how store managers spent their time: Q. Okay. And Family Dollar never did any study of the hours that the store managers were spending in the stores working, have they? A. No. Q. And Family Dollar never studied the tasks that the store managers were doing, working in the stores? A. No. Q. Family Dollar’s never studied or tried to determine how much time they spent on each task in the stores? A. There was some work done on the door to floor, you know, how much time was receiving trucks. Q. But that was not for purposes of looking at their managerial duties? A. No, sir. Q. Family Dollar’s never studied or looked into the managerial duties or the amount of time spent on managerial duties by store managers, have they? A. No, sir. Q. And Family Dollar’s never done a study or attempted to determine whether store managers are, in fact, bona fide executives, exempt from overtime, have they? A. No. *1252 Q. In fact, Family Dollar doesn’t even have a policy addressing Fair Labor Standards Act overtime requirements, does it? A. Not that I am aware of. Barkus testified that in a study of how much time it took to unload trucks and get merchandise to the floor, that the “biggest chunk of the store manager’s time was being spent on manual labor, unloading the trucks, getting it to the floor, and onto the shelves.” Barkus also testified that district managers ensure that store managers do not exceed the fixed payroll budgets assigned by corporate management. A store manager that goes over budget, by even a penny, could be fired. 12 Because store managers are under orders that overtime labor is not allowed, they are required to do any and all work, even if the payroll budget does not allocate enough hourly employees to get the job done. 13 Cuts to a store’s payroll budget necessarily reduce a store’s workforce and ensure that the salaried store manager (and not the hourly employees) makes up the difference by working more hours. 14 Almost all of the store manager’s job is standardized and controlled by superiors. Barkus confirmed that Family Dollar makes virtually no distinction between a store manager’s job duties and an assistant store manager’s job duties. 15 Plaintiffs also offered the testimony of Charles William Broome, a Senior Vice- *1253 President of Store Operations, who supervised 1,400 Family Dollar stores. Like Barkus, Broome confirmed that Family Dollar never studied whether store managers were exempt executives 16 and its exemption policy did not turn on any individual factors. It was a company wide decision that applied regardless of store size, location, sales volume, or any other individual factors: Q. You’re the official company spokesman, as you told us; correct? Now, Family Dollar has simply said all the store managers are exempt, every one of them; right? A. Yes, sir. Q. They hadn’t looked at store size? A. I — I don’t know how to answer that. Store managers have been salaried for the 29 years that I’ve been here; they were when I came. I don’t know how that was arrived at. Q. Okay. Let me ask you this. You said every store’s manager is considered exempt. And when you said “every”, you meant regardless of store size, number of employees, whether it’s rural or urban, no matter what its risk class, no matter what its sales volumes, no matter what its profits, no matter what anything; you just said they’re all exempt, didn’t you? A. All of our managers are salaried, yes, sir, in every store. Q. Doesn’t matter how many employees are in the store, they’re just all exempt? A. All the stores of Family Dollar, yes, sir. Despite his 29-year tenure with Family Dollar, Broome had no idea where the exemption decision originated: Q. Now, my question is, did you make that decision? A. No, sir. Q. Did your boss, Mr. Barkus, make that decision? A. To my knowledge, it’s been in place — it was in place when I came here 29 years ago. So— Q. Okay. So, do you know anybody that will own up to that decision; say, “that was my decision”? A. I do not. Q. Mr. Levine, has he ever told you that’s his decision? A. No, sir. Q. Can you give us any clue? And the reason I’m asking you this, I asked you this in the deposition and we’ve been asking a lot of people in depositions: Who made this decision, do you know? A. I do not. Broome acknowledged that hourly assistant managers fill in for store managers, open and close the store, can perform all managerial tasks of the store manager, and are eligible for overtime pay. Broome confirmed that a store’s payroll budget, the budgetary outlay that dictates how many employees can work in the store, is determined by corporate headquarters in conjunction with the district manager. Store managers are prohibited from exceeding the payroll budget and can be fired if they do. 17 *1254 Family Dollar’s corporate office generates a “staff scheduler” that uses the amount of money that the store may spend on labor and converts it into a document that delineates how many hours a week each employee should work and the total weekly labor hours for the store. According to the staff scheduler, each store manager is supposed to work 52 hours a week. Broome testified that, generally, store managers are expected to work between 48 to 52 hours, but that “as manager of the store, you’re required to manage the store and do whatever it takes. I don’t know that there is a specific number that’s mandated.” 18 Broome’s testimony was consistent with Family Dollar’s “Staff Schedule Frequently Asked Questions” (“FAQ”). As to whether store managers can increase or reduce associate hours, the FAQ says they may “as long as total hours & coverage for each day & week are not increased or decreased.” Similarly, as to whether the store manager can change the schedule to work “a 5/é-day workweek instead of a 5-day workweek,” the FAQ says that “[t]his change can only be made by the District Manager and should only be made as long as total hours & coverage for each day are not increased or decreased.” In other words, any flexibility store managers have in scheduling is substantially constrained by the fixed payroll budget which dictates the total labor hours. Although store managers can schedule what employees work what hours on the “weekly staff schedule” so long as the store does not exceed the payroll budget, 19 certain corporate directives further constrain store managers’ discretion, such as the prohibition on moving employee coverage from slower days (like weekdays that did not involve unloading truck shipments) to busier days. 20 Finally, the FAQ evinces Family Dollar’s strict rules against scheduling employee overtime. The FAQ indicates that store managers cannot change the schedule of assistant managers to reflect a 48 to 52 hour workweek while they are training, and that “[i]f the change is made [by the district manager] the total hours & coverage for each day should not be increased or decreased and the store payroll budget must be met. It is very important to control the use of overtime dollars.” D. District Managers Family Dollar’s 380 district managers implement and enforce these policies and *1255 procedures. Their vigorous oversight ensures that store managers comply with the operations manual’s precise dictates. The operations manual states that the district managers — not the store managers — head the “store team.” 21 District managers uniformly run their stores through strict payroll budgets, to-do lists, daily emails with instructions to store managers, telephone calls, store visits, electronic execution reports, and electronic data flowing from the store’s cash register on a real-time basis. 22 Plaintiffs’ witnesses explained how Family Dollar’s corporate office sets a fixed payroll budget for each store and how that budget results in salaried store managers working long hours each week. The district manager transmits a set payroll budget for the upcoming 13-week period to the store managers. The budget shows the store manager’s salary and a preset number of labor hours a week (to be worked by hourly employees) that must be paid for from that budget. Other than the salaried store manager, Family Dollar staffs every store with only hourly employees (either the assistant manager or sales associates). 23 District managers then closely monitor each store’s weekly payroll to ensure store managers do not allow overtime work and stay strictly within the fixed payroll budget and the total labor hours allotted. Plaintiffs introduced numerous emails confirming this fact. 24 For example, at a low volume, small store, the district manager sets the store’s payroll budget at around $1,400 per week. The average store manager’s salary of $600 per week comes out of this budget. The remaining $800 pays the hourly employees. The payroll budget is often only enough to pay one full-time hourly assistant manager and two or three hourly sales associates. Because the store is open seven days a week and store managers are not permitted to unilaterally schedule hourly staff for overtime, store managers routinely worked 60 to 70 hours a week to have enough floor coverage during the set store hours 25 and to complete the re *1256 quired manual labor. For higher volume, larger size stores, the corporate office sets a larger payroll budget, which usually covers more hourly employees (seven to ten). But larger stores have more merchandise to stock, more cartons to unload on truck day, a need for more cashiers, and more demand for cleaning. Because the payroll budget is fixed and strictly monitored, store managers at larger stores, just like those at smaller stores, routinely work 60 to 70 hours per week and spend 80 to 90% of their time on manual labor. District managers closely supervise the hiring and firing process. They interview and hire store managers and interview and approve the hiring of assistant managers. 26 Store managers initially interview assistant manager candidates and make recommendations to the district manager. The manual states that a “job offer is not to be made until ... Management has received authorization to hire.” Although store managers interview and recommend hourly associate candidates, they need district manager approval to hire them. The district manager — not the store manager — also has the authority to terminate employees. 27 The manual notes that “[pjrior to termination, Store Management should discuss the situation and its recommendation with the District Manager, or if unavailable, with the Regional Vice President.” 28 Most store managers follow corporate policy and obtain the district manager’s approval before hiring or firing hourly employees. 29 Family Dollar’s policies do not require that store managers’ hiring or firing recommendations be given any particular weight. Barkus testified that district managers can veto all store managers’ recommendations on hiring and firing. How *1257 ever, district managers frequently follow the store managers’ hiring and firing recommendations. District managers set the rate of pay for all hourly employees (assistant managers and sales associates) and must approve all pay increases. 30 District managers also evaluate hourly employee performance. 31 Only district managers have the power to close a store for bad weather. In the “Hurricane Warning Procedures” section, the manual instructs store managers that “[i]f the District manager cannot be located, contact the Regional Vice-President for recommendations regarding the course of action that should be taken.” E. Salary Compared to Hourly Wages Both parties submitted evidence documenting the average weekly salaries of all Plaintiffs and the average hourly wages of assistant managers. Plaintiffs’ evidence showed that, from 1999 to 2005, Plaintiff store managers averaged $599.71 a week in salary. 32 Despite the fact that the salary was intended to compensate a 52-hour workweek, store managers worked 60 to 70 hours a week. In other words, from 1999 to 2005, Plaintiff store managers averaged from $9.99 an hour (using Plaintiffs’ average salary figures and a 60-hour workweek) to $8.57 an hour (using Plaintiffs’ average salary figures and a 70-hour workweek). During the same years, assistant managers were paid hourly and averaged $7.60 an hour. 33 Assuming a 70-hour *1258 workweek, store managers earned, on average, roughly the same (less than a dollar or more per hour) than assistant managers. Assuming a 60-hour week, store managers earned approximately $2 more per hour than assistant managers in 1999 to 2003 and approximately $3 more per hour in 2004 to 2005. In either event, both sides presented this evidence in order to compare store manager salaries to the salaries of hourly store employees. F. Judgment/Verdict At the close of the evidence, the district court granted judgment as a matter of law to 163 of the 1,424 Plaintiff store managers, because, according to Family Dollar’s charts, these 163 did not satisfy the third requirement in the executive exemption test, i.e., that they customarily and regularly directed the work of two or more other employees, as required by 29 C.F.R. § 541.1(f) (2003), 29 C.F.R. § 541.100(a) (2006). As to the remaining Plaintiffs, the jury determined Family Dollar failed to prove they were exempt executive employees. The jury also found that Family Dollar acted willfully in denying overtime pay to all Plaintiffs. The jury awarded $1,575,932.12 in overtime pay to the 163 Plaintiffs and $17,516,071.27 in overtime pay to the remaining Plaintiffs. In calculating this overtime pay, the jury used Family Dollar’s charts that documented (1) the number of hours that each of the Plaintiffs worked per week, and (2) the amount of back pay owed per Plaintiff for the applicable period. 34 III. DECERTIFICATION A. FLSA’s Similarly Situated Requirement The FLSA authorizes collective actions against employers accused of violating the FLSA. 29 U.S.C. § 216(b). Section 216(b) provides that “[a]n action ... may be maintained against any employer ... by any one or more employees for and in behalf of himself or themselves and other employees similarly situated.” 29 U.S.C. § 216(b). Thus, to maintain a collective action under the FLSA, plaintiffs must demonstrate that they are similarly situated. See Anderson v. Cagle’s, 488 F.3d 945, 952 (11th Cir.2007). Participants in a § 216(b) collective action must affirmatively opt into the suit. 29 U.S.C. § 216(b) (“No employee shall be a party plaintiff to any such action unless *1259 he gives his consent in writing to become such a party and such consent is filed in the court in which such action is brought.”). That is, once a plaintiff files a complaint against an employer, any other similarly situated employees who want to join must affirmatively consent to be a party and file written consent with the court. Albritton v. Cagle’s, 508 F.3d 1012, 1017 (11th Cir.2007). 35 Because similarly situated employees must affirmatively opt into the litigation, the decision to certify the action, on its own, does not create a class of plaintiffs. Rather, the “existence of a collection action under § 216(b) ... depend[s] on the active participation of other plaintiffs.” Cameron-Grant v. Maxim Healthcare Servs. Inc., 347 F.3d 1240, 1249 (11th Cir. 2003) (“Under § 216(b), the action does not become a ‘collective’ action unless other plaintiffs affirmatively opt into the class by giving written and filed consent.”). 36 The benefits of a collective action “depend on employees receiving accurate and timely notice ... so that they can make informed decisions about whether to participate.” See Hoffmann-La Roche, Inc. v. Sperling, 493 U.S. 165, 170, 110 S.Ct. 482, 486, 107 L.Ed.2d 480 (1989). Therefore, the importance of certification, at the initial stage, is that it authorizes either the parties, or the court itself, to facilitate notice of the action to similarly situated employees. Hipp v. Liberty Nat’l Life Ins. Co., 252 F.3d 1208, 1218 (11th Cir. 2001). 37 After being given notice, putative class members have the opportunity to opt-in. The action proceeds throughout discovery as a representative action for those who opt-in. See id. The key to starting the motors of a collective action is a showing that there is a similarly situated group of employees. See Anderson, 488 F.3d at 953; Hipp, 252 F.3d at 1217. The FLSA itself does not define how similar the employees must be before the case may proceed as a collective action. And we have not adopted a precise definition of the term. Without defining “similarly,” we provided some guidance in Dybach v. State of Florida Department of Corrections, 942 F.2d 1562, 1567 (11th Cir.1991). There, we emphasized that before facilitating notice, a “district court should satisfy itself that there are other employees ... who desire to ‘opt-in’ and who are ‘similarly situated’ with respect to their job requirements and with regard to their pay provisions.” Id. at 1567-68. Later, in Grayson *1260 v. K Mart Corp., we instructed that under § 216(b), courts determine whether employees are similarly situated-not whether their positions are identical. 79 F.3d 1086, 1096 (11th Cir.1996). In other words, we explained what the term does not mean— not what it does. 38 Further, we review a district court’s § 216(b) certification for abuse of discretion. Hipp, 252 F.3d at 1217; Grayson, 79 F.3d at 1097. Judicial discretion in making a § 216(b) certification decision is, of course, not unbridled. Indeed, “ ‘[a] district court abuses its discretion if it applies an incorrect legal standard, follows improper procedures in making the determination, or makes findings of fact that are clearly erroneous.’ ” Anderson, 488 F.3d at 953-54 (quoting Chicago Tribune Co. v. Bridgestone/Firestone, Inc., 263 F.3d 1304, 1309 (11th Cir.2001)). The district court first must apply the proper legal standards for authorizing a § 216(b) collective action and for determining what similarly situated means. A court’s determination that the evidence shows a particular group of opt-in plaintiffs are similarly situated is a finding of fact. Anderson, 488 F.3d at 954 (affirming decision to de-certify based on conclusion “that the district court’s view of the evidence is reasonable, and its findings, therefore, are not clearly erroneous”); Hipp, 252 F.3d at 1208 (noting that decertification decision is one where the court “makes a factual determination on the similarly situated question”). We will reverse the district court’s fact-finding that Plaintiffs are similarly situated only if it is clearly erroneous- — not simply because we might have made a different call. Anderson, 488 F.3d at 953-54 (citing McMahan v. Tato, 256 F.3d 1120, 1128 (11th Cir.2001)). B. Two-Stage Procedure for Determining Certification While not requiring a rigid process for determining similarity, we have sanctioned a two-stage procedure for district courts to effectively manage FLSA collective actions in the pretrial phase. The first step of whether a collective action should be certified is the notice stage. Anderson, 488 F.3d at 952-53; Hipp, 252 F.3d at 1218. Here, a district court determines whether other similarly situated employees should be notified. A plaintiff has the burden of showing a “reasonable basis” for his claim that there are other similarly situated employees. See Anderson, 488 F.3d at 952; Grayson, 79 F.3d at 1097; Haynes v. Singer Co., 696 F.2d 884, 887 (11th Cir. 1983). 39 We have described the standard *1261 for determining similarity, at this initial stage, as “not particularly stringent,” Hipp, 252 F.3d at 1214, “fairly lenient,” id. at 1218, “flexible],” id. at 1219, “not heavy,” Grayson, 79 F.3d at 1097, and “less stringent than that for joinder under Rule 20(a) or for separate trials under 42(b),” id. at 1096. In 2007, we recounted our law and noted that at the initial stage, courts apply a “fairly lenient standard.” Anderson, 488 F.3d at 953; see Hipp, 252 F.3d at 1218. The district court’s broad discretion at the notice stage is thus constrained, to some extent, by the leniency of the standard for the exercise of that discretion. Nonetheless, there must be more than “only counsel’s unsupported assertions that FLSA violations [are] widespread and that additional plaintiffs would come from other stores.” Haynes, 696 F.2d at 887. This first step is also referred to as conditional certification since the decision may be reexamined once the case is ready for trial. Albritton, 508 F.3d at 1014 (discussing Hipp’s first stage as “conditionally certifying the collective action”); Anderson, 488 F.3d at 952 (stating district court certified collective action, “but only conditionally,” noting the possibility of later decertifying once discovery is substantially over). 40 The second stage is triggered by an employer’s motion for decertification. Anderson, 488 F.3d at 953. At this point, the district court has a much thicker record than it had at the notice stage, and can therefore make a more informed factual determination of similarity. Id. This second stage is less lenient, and the plaintiff bears a heavier burden. Id. (citing Thiessen v. Gen. Elec. Capital Corp., 267 F.3d 1095, 1103 (10th Cir.2001)). In Anderson, we again refused to draw bright lines in defining similarly, but explained that as more legally significant differences appear amongst the opt-ins, the less likely it is that the group of employees is similarly situated. Id. (“Exactly how much less lenient we need not specify, though logically the more material distinctions revealed by the evidence, the more likely the district court is to decertify the collective action.”). We also refused to “specify how plaintiffs’ burden of demonstrating that a collective action is warranted differs at the second stage.” Id. Rather, we emphasized the fact that the “ultimate decision rests largely within the district court’s discretion,” and clarified that in order to overcome the defendant’s evidence, a plaintiff must rely on more than just “allegations and affidavits.” Id. Because the second stage usually occurs just before the end of discovery, or at its close, the district court likely has a more extensive and detailed factual record. In Anderson, we also quoted approvingly of Thiessen, where the Tenth Circuit identified a number of factors that courts should consider at the second stage, such as: “(1) disparate factual and employment settings of the individual plaintiffs; (2) the various defenses available to defendants] [that] appear to be individual to each plaintiff; [and] (3) fairness and procedural considerations^]” Anderson, 488 F.3d at 953 (quoting with approval Thiessen, 267 F.3d at 1103); see also Mooney, 54 F.3d at 1213 n. 7, 1215-16. Thus, at the second stage, “although the FLSA does not require potential class members to *1262 hold identical positions, the similarities necessary to maintain a collective action under § 216(b) must extend beyond the mere facts of job duties and pay provisions” and encompass the defenses to some extent. Anderson, 488 F.3d at 953 (citation and quotation marks omitted). For example, the district court must consider whether the defenses that apply to the opt-in plaintiffs’ claims are similar to one another or whether they vary significantly. Id. at 954 n. 8 (noting that all named plaintiffs were unionized but some opt-in plaintiffs were not, making the collective bargaining agreement defense applicable to some but not all plaintiffs). But ultimately, whether a collective action is appropriate depends largely on the factual question of whether the plaintiff employees are similarly situated to one another. C. District Court’s Denial of Decertification Turning to this case and applying our circuit precedent, we conclude that Family Dollar has not shown that the district court abused its discretion in denying Family Dollar’s motion for decertification. First, the district court not only properly followed the two-stage procedure for certifying a § 216(b) collective action but also demanded even more evidence than required before certifying the case at the first notice stage. We recounted the procedural background at great length because it readily reveals that the district court proceeded very cautiously and required much more than mere allegations of similarity. The district court denied stage one certification two times, without prejudice, and continued to reexamine its decision as the parties gathered and presented additional evidence. The district court conditionally certified the collective action only after the parties filed the depositions of the named Plaintiffs and multiple affidavits and after making detailed fact-findings that Plaintiffs’ jobs were similar. 41 Subsequently, after three more years of discovery, the district court relied on a fully developed record when it denied Family Dollar’s motion for decertification, and again based its decision on detailed fact-findings. The procedural background shows that the issue of whether these 1,424 Plaintiffs were similarly situated was exhaustively litigated in the district court for over four years. At each step of the process, the district court also applied the correct legal standards under the FLSA for collection actions. Second, and more importantly, ample evidence supports the district court’s fact-findings that the Plaintiff store managers were similarly situated under § 216(b). The district court, at the second stage, had a complete and comprehensive record and found that the opt-in store managers were factually similar in a number of respects including: (1) their universal classification as store managers with the same job duties; (2) the small fraction of time they spent on managerial duties; (3) the large amount of time they spent on non-managerial duties such as stocking shelves, running the cash registers, unloading trucks, and performing janitorial work; (4) the restrictions on their power to manage stores as compared to the district manager’s sweeping managerial discretion; (5) the amount of close district manager supervision of store managers; (6) the lack *1263 of managerial discretion that Family Dollar corporate policies afforded to store managers; (7) their day-to-day responsibilities; (8) their receiving base salaries regardless of the hours worked and no overtime pay; (9) their sharing certain managerial duties with hourly employees; (10) their maintaining production and sales records; (11) their inability to authorize pay raises; (12) them power to train subordinates; (13) their restricted authority to close stores in the event of emergencies; and (14) their inability to select outside vendors without district manager approval. 42 We recognize Family Dollar’s assertion that the duties of store managers varied significantly depending on the store’s size, sales volume, region, and district. But there was scant evidence to support this argument. Rather, the bulk of evidence demonstrated that the store managers were similarly situated and even Family Dollar perceived no such distinction. Indeed, it exempted all store managers from overtime pay requirements without regard to store size, sales volume, region, district, or hiring and firing authority. 43 As to the second factor, whether there were defenses individual to each Plaintiff, Family Dollar argues the executive exemption defense is always individualized and fact specific, thereby precluding this collective action. As discussed later, applying the executive exemption is “an inherently fact-based inquiry” that depends on the many details of the particular job duties and actual work performed by the employee seeking overtime pay. See Rodriguez v. Farm Stores Grocery, Inc., 518 F.3d 1259, 1263 (11th Cir.2008). But Family Dollar ignores the overwhelming evidence showing that the Plaintiffs, as a group, shared a number of factual details with respect to their job duties and day-to-day work. Just because the inquiry is fact-intensive does not preclude a collective action where plaintiffs share common job traits. Given the volume of evidence showing the store managers were similarly situated, and the fact that Family Dollar applied the executive exemption across-the-board to every store manager — no matter the size, region, or sales volume of the store — Family Dollar has not shown clear error in the district court’s finding that its defenses were not so individually tailored to each Plaintiff as to make this collective action unwarranted or unmanageable. 44 *1264 As to the third factor, fairness and procedural considerations, we reject Family Dollar’s contention that given the size of the class, the individualized application of the exemption defense, and the court’s decision to allow representative testimony at trial, any collective action would be inherently unfair. There is nothing inherently unfair about collectively litigating an affirmative executive-exemption defense where the district court has made well-supported and detailed findings with respect to similarity. Indeed, the more similar the employees, the less likely that collectively litigating the executive-exemption issue can be fundamentally unfair. And to repeat, there was abundant evidence that Plaintiffs’ actual jobs were the same, or, at a minimum, substantially similar. 45 In addition, Plaintiffs’ evidence established that Family Dollar uniformly exempted all store managers from overtime pay requirements, and its exemption decision did not turn on any individualized factors. Not one. There is nothing unfair about litigating a single corporate decision in a single collective action, especially where there is robust evidence that store managers perform uniform, cookie-cutter tasks mandated by a one-size-fits-all corporate manual. 46 Addressing whether Plaintiffs’ claims could be tried fairly as a collective action also requires looking to the purposes of § 216(b) actions under the FLSA: (1) reducing the burden on plaintiffs through the pooling of resources, and (2) efficiently resolving common issues of law and fact that arise from the same illegal conduct. See Hoffmann-La Roche, Inc. v. Sperling, 493 U.S. 165, 170, 110 S.Ct. 482, 107 L.Ed.2d 480 (1989) (noting a collective *1265 action affords plaintiffs the “advantage of lower individual costs to vindicate rights by the pooling of resources” and “[t]he judicial system benefits by efficient resolution in one proceeding of common issues of law and fact”). We also bear in mind that the FLSA is a remedial statute that should be liberally construed. Prickett v. DeKalb County, 349 F.3d 1294, 1296 (11th Cir. 2003) (“FLSA is a remedial statute that has been construed liberally to apply to the furthest reaches consistent with congressional direction.” (quotation marks omitted)). Oddly, the thrust of Family Dollar’s fairness argument butts up against the purpose of a collective action-to efficiently resolve a large number of plaintiffs’ claims. 47 Therefore, generally speaking, the size of an FLSA collective action does not, on its own, compel the conclusion that a decision to collectively litigate a case is inherently unfair. Furthermore, Family Dollar’s decertifi-cation argument is, at root, a claim that the district court’s subsequent use of representative testimony during the actual trial was inherently unfair. As discussed later, Family Dollar’s representative — testimony assertion — that the district court allowed the jury to decide 1,424 claims based on the testimony of only seven named Plaintiffs — is not supported by the record. In any event, we examine the evidence before the court when it heard Family Dollar’s motion to decertify. See Haynes, 696 F.2d at 887 (“Our review of that decision must be premised upon the evidence that was before the district court at that time.”). We are persuaded that, before trial, fairness considerations militated in favor of allowing this overtime-pay action to proceed in a collective forum. Given the substantial similarity of the class members’ jobs and uniform corporate treatment of the store managers, it would not serve the interest of judicial economy to require these overtime-pay claims to be adjudicated in 1,424 individual trials. Based on the record in this case, we cannot say the district court abused its discretion in denying Family Dollar’s motion for decertification. IV. EXECUTIVE EXEMPTION DEFENSE A. FLSA’s Executive Exemption The FLSA requires that employers pay their employees time and a half for hours an employee works in excess of a 40-hour workweek. 29 U.S.C. § 207(a)(1); Alvarez Perez v. Sanford-Orlando Kennel Club, Inc., 515 F.3d 1150, 1156 (11th Cir.2008); Allen v. Bd. of Pub. Educ., 495 F.3d 1306, 1314 (11th Cir.2007). But there are exemptions to this requirement. Alvarez Perez, 515 F.3d at 1156. The exemption at issue here, the executive exemption, provides that the FLSA’s requirements “shall not apply with respect to ... any employee employed in a bona fide executive ... capacity.” 29 U.S.C. § 213(a)(1). The Department of Labor’s (“DOL”) regulations interpret this defense. Because Plaintiffs’ claims span from 1999 until 2005, two sets of DOL regulations ap *1266 ply. The regulations that were in effect prior to August 23, 2004 (the “old regulations”) contain a short test that defines the phrase “employee employed in a bona fide executive capacity.” 48 29 C.F.R. § 541.1 (2003) (ellipsis omitted). This short test has three requirements: (1) an employee “is compensated on a salary basis at a rate of not less than $250 per week,” (2) his “primary duty consists of the management of the enterprise in which the employee is employed or of a customarily recognized department or subdivision thereof,” and (3) his work “includes the customary and regular direction of the work of two or more other employees.” Id. (2003); Rodriguez, 518 F.3d at 1263; Brock v. Norman’s Country Mkt., Inc., 835 F.2d 823, 825 (11th Cir.1988). 49 After August 23, 2004, the new regulations apply and add a fourth requirement. To establish an employee is a bona fide executive, an employer must show: (1) the employee is “[compensated on a salary basis at a rate of not less than $455 per week”; (2) the employee’s “primary duty is management of the enterprise in which the employee is employed or of a customarily recognized department or subdivision thereof’; (3) the employee “customarily and regularly directs the work of two or more other employees”; and (4) the employee “has the authority to hire or fire other employees or whose suggestions and recommendations as to the hiring, firing, advancement, promotion or any other change of status of other employees are given particular weight.” 29 C.F.R. § 541.100(a) (2006). 50 The parties agree that the first element of the executive exemption test — the amount of salary — is met. But they hotly dispute the second element — whether the store managers’ primary duty is management. Thus, we examine the second element. B. Primary Duty Is Management Both regulations require that the employee’s primary duty is management. *1267 The old regulations give these examples of managerial tasks: Interviewing, selecting, and training of employees; setting and adjusting their rates of pay and hours of work; directing their work; maintaining their production or sales records for use in supervision or control; appraising their productivity and efficiency for the purpose of recommending promotions or other changes in their status; handling their complaints and grievances and disciplining them when necessary; planning the work; determining the techniques to be used; apportioning the work among the workers; determining the type of materials, supplies, machinery or tools to be used or merchandise to be bought, stocked and sold; controlling the flow and distribution of materials or merchandise and supplies; providing for the safety of the men and the property. 29 C.F.R. § 541.102(b) (2003). 51 The new regulations are similar and explain that generally, management includes, but is not limited to, the same conduct listed in the old regulations, but adds these examples of managerial tasks: “planning and controlling the budget; and monitoring or implementing legal compliance measures.” 52 29 C.F.R. § 541.102 (2006). The old and new regulations do not define primary duty. Both indicate the answer to the primary duty question “must be based on all the facts in a particular case.” 29 C.F.R. § 541.700(a) (2006); 29 C.F.R. § 541.103 (2003). Both regulations identify factors to consider when determining whether an employee’s primary duty is managerial. See 29 C.F.R. § 541.700(a) (2006); 29 C.F.R. § 541.103 (2003). The old regulations list these five factors: (1) “[t]he amount of time spent in the performance of the managerial duties”; (2) “the relative importance of the managerial duties as compared with other types of duties”; (3) “the frequency with which the employee exercises discretionary powers”; (4) “his relative freedom from supervision”; and (5) “the relationship between [the employee’s] salary and the wages paid other employees for the kind of nonexempt work performed by the supervisor.” 29 C.F.R. § 541.103 (2003); see also Rodriguez, 518 F.3d at 1264 (listing factors in § 541.103 to be analyzed in determining whether an employee’s primary duty is management). The old regulations explain: “In the ordinary case it may be taken as a good rule of thumb that primary duty means the major part, or over 50 percent, of the employee’s time. Thus, an employee who spends over 50 percent of his time in management would have *1268 management as his primary duty.” 29 C.F.R. § 541.103 (2003). The regulations then hedge a bit, and state that “[t]ime alone ... is not the sole test .... ” Id. (2003). Where the employee does not spend over 50% of his time on management, “he might nevertheless have management as his primary duty if the other pertinent factors support such a conclusion.” Id. (2003). The new regulations make a stronger effort to define primary duty, stating that “[t]he term ‘primary duty’ means the principal, main, major or most important duty that the employee performs.” 29 C.F.R. § 541.700(a) (2006). The new regulations also add that this determination is to be made “with the major emphasis on the character of the employee’s job as a whole.” Id. (2006). The new regulations explicitly reference the same factors with one exception. The third factor — “the frequency with which the employee exercises discretionary powers” — has been deleted. 53 Compare id. (2006), with 29 C.F.R. § 541.103 (2003). The new regulations, like the old, expand upon the time-spent-on-exempt-work factor. The new regulations state that this factor “can be a useful guide in determining whether exempt work is the primary duty of an employee” and that “employees who spend more than 50 percent of their time performing exempt work will generally satisfy the primary duty requirement.” 29 C.F.R. § 541.700(b) (2006). As in the old ones, the new regulations specify that “[t]ime alone, however, is not the sole test” and thus “[ejmployees who do not spend more than 50 percent of their time performing exempt duties may nonetheless meet the primary duty requirement if the other factors support such a conclusion.” Id. (2006). The new regulations also clarify that “[cjoncurrent performance of exempt and nonexempt work does not disqualify an employee from the executive exemption if the requirements of § 541.100 are otherwise met.” Id. § 541.106(a) (2006). In other words, an employee’s performance of nonexempt work does not preclude the exemption if the employee’s primary duty remains management. 54 Similarly, an employee whose primary duty is to perform nonexempt work does not become exempt merely because she has some responsibility for occasionally directing the work of nonexempt employees. 55 “Whether an em *1269 ployee meets the requirements of § 541.100 when the employee performs concurrent duties is determined on a case-by-case basis” and based on the factors already set forth in § 541.700(a) as to the primary duty question. Id. (2006). We turn to Family Dollar’s argument that the primary duty of its store managers is managerial, triggering the FLSA’s executive exemption, and that the court erred in denying Family Dollar’s motion for judgment as a matter of law. C. Family Dollar’s Motion for Judgment as a Matter of Law Family Dollar bears the burden of proving its executive exemption affirmative defense. Alvarez Perez, 515 F.3d at 1156; Brock, 835 F.2d at 826. This Court has recognized the “Supreme Court’s admonition that courts closely circumscribe the FLSA’s exceptions.” Nicholson v. World Bus. Network, Inc., 105 F.3d 1361, 1364 (11th Cir.1997). And the exemption “is to be applied only to those clearly and unmistakably within the terms and spirit of the exemption.” Brock, 835 F.2d at 826 (quotation marks omitted). Therefore, we narrowly construe exemptions to the FLSA overtime requirement. Alvarez Perez, 515 F.3d at 1156 (stating “exemptions are to be construed narrowly” (quotation marks omitted)); Nicholson, 105 F.3d at 1364. We have rejected a “categorical approach” to deciding whether an employee is an exempt executive. Rodriguez, 518 F.3d at 1264. Instead, we have noted the “necessarily fact-intensive nature of the primary duty inquiry,” that “the answer is in the details,” and that “[w]here an issue turns on the particular facts and circumstances of a case, it is not unusual for there to be evidence on both sides of the question, with the result hanging in the balance.” Id. And “[t]he result reached must be left intact if there is evidence from which the decision maker, the jury in this instance, reasonably could have resolved the matter the way it did.” Id. Here, the trial evidence was legally sufficient for a reasonable jury to find that Family Dollar failed to meet its burden of proving that its store managers’ primary duty was management. Because that evidence is detailed above, we do not recount it but focus on the factors in the primary duty inquiry. 56 As to the time-spent-on-exempt-work factor, the overwhelming evidence at trial showed Plaintiff store managers spent 80 to 90% of their time performing nonexempt, manual labor, such as stocking shelves, running the cash registers, unloading trucks, and cleaning the parking lots, floors, and bathrooms. Conversely, Plaintiff store managers spent only 10 to 20% of their time performing exempt work, a far cry from the DOL’s 50% guideline for management tasks. See 29 C.F.R. § 541.700(b) (2006); 29 C.F.R. § 541.103 (2003). Family Dollar did not present evidence to the contrary. See Allen, 495 F.3d at 1315 (“The employer is in a superior position to know and produce the most *1270 probative facts concerning the nature and amount of work performed .... ”). In fact, Family Dollar executives Barkus and Broome testified that Family Dollar never studied what store managers actually did on a day-to-day basis or the amount of time store managers spent on managerial duties. We recognize that the amount of time spent performing exempt tasks is not dispositive of the primary duty issue. But substantial evidence about the four other factors also supports the jury’s verdict here. As to the relative importance of store managers’ managerial duties compared with their nonexempt duties, this factor weighs in favor of the jury’s finding that store managers are not exempt executives. Admittedly, the store managers’ job description includes managerial duties. But Family Dollar’s job description of the store managers’ “Essential Job Functions” provides that store managers must do the same work as stock clerks and cashiers. Store managers must work their store’s preassigned merchandise delivery day, known as “truck day.” Barkus acknowledged that store managers spent their delivery-day time doing manual labor. Rather than treat these manual tasks as an incidental part of a managerial job, Family Dollar describes them as essential. A large amount of manual labor by store managers was a key to Family Dollar’s business model given each store’s limited payroll budget and the large amount of manual labor that had to be performed. The jury was free to weigh the relative importance of the store managers’ managerial and non-managerial duties, but ample evidence supported a finding that the non-managerial tasks not only consumed 90% of a store manager’s time but were of equal or greater importance to a store’s functioning and success. The third factor in the old regulations— the frequency with which the employee exercises discretionary powers — also supports the jury’s verdict. 57 There was overwhelming evidence that store managers spent only 10 to 20% of their time on exempt (i.e., managerial) work. Plaintiffs presented evidence that store managers rarely exercised discretion because either the operations manuals or the district managers’ directives controlled virtually every aspect of a store’s day-to-day operations. The manuals and other corporate directives micro-managed the days and hours of store operations, the number of key sets for each store, who may possess the key sets, entire store layouts, the selection, presentation, and pricing of merchandise, promotions, payroll budgets, and staffing levels. The manuals even instruct store managers on the smallest details, such as how to arrange clip boards, what items go in each of the four drawers of the single file cabinet, and how to remove spots and chewing gum from store mats. The few decisions not mandated by the manuals and corporate headquarters are vested in the district manager. These decisions include the power to change store hours, close for bad weather, approve changes to store layouts, establish all employees’ initial rates of pay, approve all pay raises, set payroll budgets, control the *1271 total labor hours allocated to each store, approve the hiring and firing of assistant managers, and even approve the use of appliances such as coffee pots. Even when a store manager exercised discretion in scheduling employees for the week, she did so within the strict constraints of mandatory store hours, a limited payroll budget, a prohibition on overtime work by hourly employees, and a staff scheduler. This evidence supports a reasonable jury finding that Family Dollar’s store managers had few, and infrequently exercised, discretionary powers. As to the store managers’ relative freedom from supervision, this factor likewise favors Plaintiffs. The evidence showed that district managers (1) supervised 10 to 30 stores, (2) headed the store team, (3) were responsible for enforcing the detailed store operating policies, (4) closely reviewed each store’s inventory orders and net sales figures, (5) closely monitored each store’s weekly payroll, (6) controlled employee hourly rates and pay raises, (7) routinely sent to-do lists and emails with instructions to store managers, (8) closely supervised the selection, pricing, sales, displays, and ordering of merchandise, and (9) closely supervised every aspect of store operations. Store managers had little freedom from direct supervision. Indeed, ample evidence showed that the combination of sweeping corporate micro-management, close district manager oversight, and fixed payroll budgets left store managers little choice in how to manage their stores and with the primary duty of performing manual, not managerial, tasks. 58 As to the last factor- — the relationship between the store managers’ salary and other wages for nonexempt work — the parties submitted evidence documenting the store managers’ average salaries and the assistant managers’ average hourly wages from 1999 to 2005. Using a 70-hour workweek, store managers earned, on average, roughly the same (less than a dollar or more per hour) as hourly assistant managers. Using a 60-hour workweek, store managers earned approximately two or three dollars more per hour than hourly assistant managers. The jury was entitled to consider these salary figures, along with the fact that store managers performed nonexempt work 80 to 90% of the time. Given the relatively small difference between the store managers’ and assistant managers’ hourly rates, it was within the jury’s province to conclude that this factor either did not weigh in Family Dollar’s favor or at least did not outweigh the other factors in Plaintiffs’ favor. In sum, there was legally sufficient evidence for the jury, after considering all of the evidence and weighing the relevant factors, 59 to have reasonably determined that Family Dollar failed to meet its burden of proving that Plaintiff store managers’ primary duty was management. D. Other Circuits’ Cases Despite these factors, Family Dollar insists its store managers were “in charge” of the store, and therefore, exempt as a matter of law. Family Dollar cites several cases concluding that managers of a freestanding store or restaurant were exempt executives as a matter of law. However, the courts made that decision only after examining the factual details of the employees’ duties and actual work. As we *1272 stated in Rodriguez, such a “particularized approach is consistent with the DOL regulation, which provides that the ‘determination of whether an employee has management as his primary duty must be based on all the facts in a particular case.’ ” 518 F.3d at 1264 (quoting 29 C.F.R. § 541.103 (2002)); see 29 C.F.R. § 541.700(a) (2006) (stating the primary duty question “must be based on all the facts in a particular case”); 29 C.F.R. § 541.103 (2003) (same). In answering the primary duty inquiry, courts do not “simply slap[ ] on a talismanic phrase.” Rodriguez, 518 F.3d at 1264. Family Dollar’s “in charge” label strikes us as a way to bypass a meaningful application of the fact-intensive factors. As in Rodriguez, we reject that “categorical approach.” Id. Moreover, Family Dollar’s cases had materially dissimilar facts and did not involve the combination of (1) store managers performing as high a percentage of nonexempt work, (2) the same severe degree of restriction on store managers’ discretion by corporate policy, and (3) oversight as strict and involved as district managers’ in this case. For starters, none of the circuit cases dealt with store managers that spent 80 to 90% of the time performing manual labor. For example, in Thomas v. Speedway SuperAmerica, LLC, 506 F.3d 496, 499 (6th Cir.2007), a store manager spent 60% of her time performing non-managerial duties. 60 See also Baldwin v. Trailer Inns, Inc., 266 F.3d 1104, 1108-09, 1114 (9th Cir.2001) (accepting on summary judgment plaintiffs’ supported assertion they spent “more than fifty percent of the time” on nonexempt manual tasks but without providing a concrete figure). 61 Family Dollar’s cases are also distinguishable in that they give less weight to the plaintiffs’ estimates of time spent performing nonexempt work because the plaintiffs in those cases performed exempt and nonexempt work concurrently. See Thomas, 506 F.3d at 504; Donovan v. Burger King Corp. (“Burger King /”), 672 F.2d 221, 226 (1st Cir.1982); Baldwin, 266 F.3d at 1114. The evidence, in the light most favorable to the Plaintiffs here, did not show they performed managerial and non-managerial tasks concurrently. Rather, there was evidence that, by and large, the store managers performed most managerial tasks before the store opened and after it closed. The amount of manual labor overwhelmed their capacity to perform managerial duties concurrently during store hours. Other evidence showed *1273 the nature of the manual labor prevented store managers from performing managerial duties concurrently. For example, a store manager unloading a truck and stocking the storeroom was not concurrently supervising the cashier out front. The jury may well have given more weight to the Plaintiffs’ evidence that they spent 80 to 90% of their time solely on nonexempt work. Even the retail-chain cases with standardized instructions did not involve fact patterns with the same level of corporate directives or district managers that constrained the powers of the employees-in-question in quite the same way. For example, in Burger King I, 672 F.2d at 223, and Donovan v. Burger King Corp. (“Burger King II”), 675 F.2d 516, 517, 521-22 (2d Cir.1982), the assistant managers retained discretion over a number of operational decisions, and nothing suggested the Burger King restaurant manager, the position directly above the assistant manager, had oversight powers comparable to the ones exercised by Family Dollar district managers. The Second Circuit described restaurant managers as “available by phone” and “available for advice”-not as overarching remote micro-managers. See Burger King II, 675 F.2d at 522. Likewise, in Murray v. Stuckey’s, Inc., 939 F.2d 614, 616 (8th Cir.1991), the court’s description of regional-manager control pales in comparison to the rigid directives and supervision that Plaintiffs presented at trial. Id. For Family Dollar store managers, there was evidence that showed district managers and corporate headquarters made the vast majority of day-to-day decisions, and store managers had little discretion. In any event, our affirmance of the jury’s verdict here is based on a fact-intensive application of the factors espoused in the regulations, and not on a categorical approach of whether a particular employee is “in charge.” More importantly, while there was “evidence on both sides of the question,” the “jury[,j in this instance, reasonably could have resolved the matter the way it did.” Rodriguez, 518 F.3d at 1264. “The issue is not whether the evidence was sufficient for [Family Dollar] to have won, but whether the evidence was sufficient for it to have lost. It was.” Id. at 1264-65. 62 E. 16S Individual Plaintiffs Granted Judgment on Executive Exemption Defense Family Dollar also appeals the district court’s decision to grant judgment on the executive exemption defense as a matter of law to 163 of the 1,424 individual Plaintiffs. The court concluded that Family Dollar failed to prove these 163 Plaintiffs satisfied the third part of the executive exemption test, i.e., that they customarily and regularly directed the work of two or more other employees. 29 C.F.R. § 541.100(a) (2006) (identifying third part of executive exemption test as whether the employee “customarily and regularly directs the work of two or more other employees”); 29 C.F.R. § 541.1 *1274 (2003) (requiring that an exempt employee’s work “includes the customary and regular direction of the work of two or more other employees”). As we explain below, Family Dollar has shown no reversible error on this issue. Both regulations provide that customarily and regularly means a frequency that “must be greater than occasional but which, of course, may be less than constant.” 63 29 C.F.R. § 541.701 (2006); 29 C.F.R. § 541.107(b) (2003). The new regulations add that “[t]asks or work performed ‘customarily and regularly’ includes work normally and recurrently performed every workweek; it does not include isolated or one-time tasks.” 29 C.F.R. § 541.701 (2006). Both regulations define “two or more other employees” as either two full-time workers or their equivalent. 29 C.F.R. § 541.104(a) (2006); 29 C.F.R. § 541.105(a) (2003). As to equivalency, an employee may supervise one full-time employee and two part-time employees. See 29 C.F.R. § 541.105(a) (2003). “For example, if the ‘executive’ supervises one full-time and two part-time employees of whom one works morning[s] and one, afternoons; or four part-time employees, two of whom work mornings and two afternoons, this requirement would be met.” Id. The new regulations express the same sentiment. See 29 C.F.R. § 541.104(a) (2006) (“One full-time and two half-time employees, for example, are equivalent to two full-time employees. Four half-time employees are also equivalent.”). However, neither set of regulations defines “full-time.” To prove its store managers customarily and regularly directed the work of two or more employees, Family Dollar introduced payroll records and easy-to-digest exhibits summarizing that data. For example, Exhibit 1742C provided data about each of the 1,424 Plaintiffs including (1) their full names, (2) social security numbers, (3) the number of weeks each worked as a store manager (under a row heading entitled “# weeks supervised”), (4) the number of weeks each store manager had two or more full-time hourly employees working in the store, and (5) what percentage of time each store manager had two or more full-time hourly employees. Exhibit 1742C assumed that a full-time employee works 40 hours a week and that full-time meant 80 hours of employee work per week through any combination of employees. 64 In applying the requirement that an employee “customarily and regularly directs the work of two or more other employees,” the district court examined whether store managers supervised 80 subordinate hours of employee work per week at least 80% of the time. The district court did not require the store managers to be present when hourly employees worked those 80 hours — only that the store employ 80 hours of subordinate employee labor 80% of the time. Using Family Dollar’s Exhib *1275 it 1742C, the district court granted judgment as a matter of law to 163 salaried Plaintiff store managers whose stores did not meet that requirement. Family Dollar argues the court should have used Family Dollar’s internal definition of full-time as a 30-hour workweek. Although the preamble to the new regulations states that the DOL declines to clarify the meaning of “full-time,” it “stands by its current interpretation that an exempt supervisor generally must direct a total of 80 employee-hours of work each week.” 65 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 69 Fed.Reg. 22,122, 22,135 (Apr. 23, 2004). The preamble also states that “circumstances might justify lower standards. For example, firms in some industries have standard workweeks of 37/6 hours or 35 hours for their full-time employees. In such cases, supervision of employees working a total of 70 or 75 hours in a workweek will constitute the equivalent of two full-time employees.” Id. Notably, the preamble does not suggest a workweek as short as 30 hours counts as “full-time” under the FLSA, much less 60 labor hours substituting for 80 hours. Further, Family Dollar’s brief points to no evidence in the trial record suggesting that the industry standard in its line of retail business is a 30-hour workweek. While there may be instances where a deviation from the 40-hour workweek is appropriate, we cannot say that the district court, based on this record, erred in adopting 80 hours as constituting two full-time employees or their equivalent in this case. Although Family Dollar also criticizes the 80% threshold, Family Dollar does not argue that the court’s effort to quantify the customary — regular requirement as a percentage of time was error or that the customary — regular issue should have gone to the jury. Rather, Family Dollar primarily challenges the district court’s last step in the calculation of whether Plaintiffs met the 80% threshold. As we noted above, the district court authorized Family Dollar to count every week that the stores in question had 80 labor hours worked by the hourly employees, and it did not require the store managers to be physically present at the store during those 80 hours for those weeks to count towards the 80% mark. 66 Family Dollar does not challenge that part of the calculation. 67 Nor does it challenge the accuracy of the calculation as to the number of weeks each salaried store manager worked, the number of weeks hourly em *1276 ployees worked, or the percentage of time each individual store manager directed the work of two or more full-time employees or their equivalent. Indeed, the data are from Family Dollar’s own Exhibit 1742C. The calculations showed, for example, that 589 stores had 80 labor hours 100% of the time, and 290 more stores had 80 labor hours 90% of the time. But Exhibit 1742C also showed that 163 stores had 80 labor hours less than 80% of the time in their stores (and of those, some were well below 50% and several at 0%). Family Dollar contends that the court should have used the average percentage of all 1,424 store managers as a group. In other words, its argument is that, on average, store managers (as a group) had 80 labor hours in their stores 93% of the time. Family Dollar arrives at this conclusion by dividing the number of weeks that Plaintiffs supervised two or more employees for 80 hours (61,481) by the total number of weeks that Plaintiffs worked at Family Dollar (66,097). The total percentage: 93%. While averaging the number of weeks that employees meet the criteria may suffice in some collective cases, the court here enjoyed more precise data that showed (1) the percentage for each of the 1,424 store managers individually and (2) that 163 of the Plaintiffs failed to supervise two or more employees 80% of the time. Family Dollar cites no authority, and presents us with no principled basis, for determining that the court’s use of this more precise evidence was reversible error. The parties cite two other circuit court cases addressing this issue. In Secretary of Labor v. Daylight Dairy Prods., Inc., the First Circuit concluded that employees who do not direct work 76% of the time “fall[ ] short” of “regular and customary” supervision of 80 hours of work. 779 F.2d 784, 788 (1st Cir.1985). Although Daylight Dairy failed to decide what precise figure constitutes customary and regular direction, it determined that 76% was insufficient. In the Eighth Circuit’s Murray decision, the “store managers supervised at least two or more employees who worked eighty hours per week 98.2% of the time.” 50 F.3d at 567-68. The Eighth Circuit concluded: “That is ‘customarily and regularly’ by any definition.” Id. Although the court expressed its disagreement with Daylight Dairy, id. at 568, it similarly failed to define a minimum threshold of precisely what “customary and regular” means. We likewise do not draw a bright-line rule but examine only whether Family Dollar has shown reversible error in the court’s use of Family Dollar’s Exhibit 1742C as a basis for granting judgment in favor of the 163 Plaintiff store managers. It has not. V. REPRESENTATIVE TESTIMONY Family Dollar’s challenge to the use of representative testimony proceeds as follows: Seven Plaintiffs testified. There are 1,424 Plaintiffs. Therefore, the verdict is based on only the representative testimony of less than 1% of the total number of Plaintiffs. Family Dollar argues that this was simply too small a sample size of testifying Plaintiffs, and therefore the jury verdict is unreliable and should be set aside. Family Dollar’s depiction of the trial is belied by the record. First, Plaintiffs did not use representative testimony to prove its prima facie case. Instead, Plaintiffs relied on Family Dollar’s thorough payroll records for each of the 1,424 Plaintiffs to show (1) when each employee worked, (2) how many actual hours they worked, (3) how much they were paid, and (4) that they never received overtime pay. Rather than contesting Plaintiffs’ prima facie *1277 case, 68 Family Dollar focused on its executive exemption defense at trial. Second, Family Dollar’s claim, stripped of its hyperbole, is reduced to an objection that not enough Plaintiffs testified to ensure a reliable verdict on whether the executive exemption defense applied. But the jury’s verdict as to that defense was not based on the testimony of just seven Plaintiffs. Instead, the parties presented an abundance of trial evidence about the executive exemption issue, including (1) a vast array of corporate manuals; (2) testimony from 39 witnesses including Family Dollar executives, district managers who ran the operations of 134 stores, and store managers who worked at a total of 50 different stores; (3) detailed charts summarizing wages and hours; and (4) a wealth of exhibits including emails, internal Family Dollar correspondence, payroll budgets, and in-store schematics. If one factors in that Broome and Barkus oversaw thousands of stores, the witnesses go from representing hundreds of stores to thousands. In addition to the large quantity of testimonial evidence, the non-testimonial evidence was equally high in quality and largely comprised of Family Dollar’s corporate records. The jury’s verdict is well-supported not simply by “representative testimony,” but rather by a volume of good old-fashioned direct evidence. 69 Family Dollar’s trial conduct is also revealing. It actually opposed the introduction of more witness testimony from Plaintiff store managers. After Family Dollar presented the deposition testimony of 12 opt-in store managers, Plaintiffs attempted to introduce into evidence the deposition testimony of 238 more opt-in store managers. But Family Dollar objected, 70 and the *1278 district court sustained the objection. 71 Family Dollar cannot validly complain about the number of testifying plaintiffs when it successfully objected to Plaintiffs’ attempt to present the testimony of almost 20 times as many Plaintiff store managers. Although Family Dollar itself had the opportunity to present a great deal more testimony from Plaintiff store managers, or its own district managers, it chose not to. Indeed, Family Dollar used only 10 of its allotted 40 hours for its defense, even though it bore the burden of proving the executive exemption defense. 72 This leads us to a third flaw in Family Dollar’s argument. Plaintiffs did not shoulder the burden of proof on the executive exemption defense. Family Dollar did. See Atlanta Prof'l Firefighters Union, 920 F.2d at 804. Thus, Family Dollar cannot rely on an insufficient number of witnesses being called by the Plaintiffs to meet Family Dollar’s burden of proof on its own affirmative defense. Fourth, Family Dollar relies on two FLSA decisions involving representative testimony, but they do not help Family Dollar. See Reich v. S. Md. Hosp., Inc., 43 F.3d 949, 951-52 (4th Cir.1995); Sec’y of Labor v. DeSisto, 929 F.2d 789, 792-96 (1st Cir.1991). Since these cases are part of a line of circuit cases dating back to Anderson v. Mount Clemens Pottery, 328 U.S. 680, 66 S.Ct. 1187, 90 L.Ed. 1515 (1946), 73 a few words about Mt. Clemens are in order first. In Mt. Clemens, the Supreme Court authorized a burden-shifting scheme designed to facilitate the ability of plaintiffs to prove an FLSA violation where the employer failed to maintain proper records (such as how many hours its employees worked and the amount of pay). Id. at 686-88, 66 S.Ct. at 1192. To prevent workers from being penalized by the employer’s failure to keep adequate records, the Supreme Court provided that plaintiffs could meet their burden of proof so long as they “prove[ ] that [they have] in fact performed work for which [they were] improperly compensated” and “produee[ ] sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference.” Id. at 687, 66 S.Ct. at 1192. Although Mt. Clemens never used the term “representative testimony,” subsequent courts have interpreted it to authorize some employees to testify about the number of hours they worked and how much they were paid so that other non-testifying plaintiffs could show the same *1279 thing by inference. See, e.g., Reich, 43 F.3d at 951 (“Under Mt Clemens, the Secretary can present testimony from representative employees as part of his proof of the prima facie case.”); Donovan, 780 F.2d at 1116 (“There is no requirement that to establish a Mt Clemens pattern or practice, testimony must refer to all non-testifying employees. Such a requirement would thwart the purposes of the sort of representational testimony clearly contemplated by Mt Clemens.”); DeSisto, 929 F.2d at 792 (interpreting Mt Clemens burden-shifting and noting that the plaintiff “can rely on testimony and evidence from representative employees to meet the initial burden of proof requirement”); McLaughlin v. Ho Fat Seto, 850 F.2d 586, 589 (9th Cir.1988) (“We hold that the Mt Clemens Pottery standard allows district courts to award back wages under the FLSA to non-testifying employees based upon the fairly representative testimony of other employees.”). As a result, most of the circuit cases addressing how many FLSA plaintiffs need to testify before their testimony can be considered representative of the group involve an employer’s failure to keep adequate records and the plaintiffs’ use of the Mt Clemens burden-shifting scheme. See, e.g., McLaughlin, 850 F.2d at 589; Donovan v. Bel-Loc Diner, Inc., 780 F.2d 1113, 1115-16 (4th Cir.1985), disapproved of on other grounds, McLaughlin v. Richland Shoe Co., 486 U.S. 128, 108 S.Ct. 1677, 100 L.Ed.2d 115 (1988); Brennan v. Gen. Motors Acceptance Corp., 482 F.2d 825, 829 (5th Cir.1973). Family Dollar relies on two of these kinds of cases to support its argument that, statistically, not enough plaintiffs testified here. See Reich, 43 F.3d at 951-52 (reversing district court because testimony from 58 employees was insufficient to represent 3,368 employees); DeSisto, 929 F.2d at 792-96 (reversing district court for allowing one employee to represent 244 employees at trial). This line of cases does not help Family Dollar. Here, Family Dollar adequately maintained its records. Indeed, Plaintiffs relied on those records extensively. Therefore, the Mt. Clemens burden-shifting analysis does not apply. Furthermore, the question in these burden-shifting cases is whether the plaintiffs showed the amount and extent of the work performed as a matter of just and reasonable inference. In such a context, it makes sense to examine whether there is, statistically speaking, enough evidence to support the inference, and to shift the burden of proof on an element of the plaintiffs’ case (the number of hours worked) to the employer. See DeSisto, 929 F.2d at 794 (“The evidence was simply inadequate to give rise to a ‘just and reasonable inference’ as to the amount and extent of undercompensat-ed work.”). Where employer payroll records are inadequate, litigants can only approximate the number of hours worked and the amount of pay due. In such cases, the answer requires a numerical estimate of hours and pay. In contrast, the Plaintiffs here relied on Family Dollar’s extensive payroll records that broke down, week-by-week, how many hours each of the 1,424 store managers worked. Reich and DeSisto are nothing like this case. Here, there was no need for such numerical approximation. If anything, the Mt. Clemens line of cases affirms the general rule that not all employees have to testify to prove overtime violations. See DeSisto, 929 F.2d at 793 (indicating that, generally speaking, employees who perform “substantially similar work” may testify on behalf of their counterparts); Reich v. S. New England Telecomms. Corp., 121 F.3d 58, 63 (2d Cir.1997) (stating “there is no bright line formulation that mandates reversal when *1280 the sample is below a percentage threshold. It is axiomatic that the weight to be accorded evidence is a function not of quantity but of quality.”); see also Brock v. Norman’s Country Mkt., Inc., 835 F.2d 823, 828 (11th Cir.1988) (“[I]t is clear that each employee need not testify in order to make out a prima facie case of the number of hours worked as a matter of ‘just and reasonable inference.’ ”) (quoting Donovan v. New Floridian Hotel, Inc., 676 F.2d 468, 472 (11th Cir.1982)). 74 We reject Family Dollar’s argument that the executive exemption defense is so individualized that the testifying Plaintiffs did not fairly represent the non-testifying Plaintiffs. For the same reasons that the court did not err in determining that the Plaintiffs were similarly situated enough to maintain a collective action, it did not err in determining that the Plaintiffs were similarly situated enough to testify as representatives of one another. In any event, the only issue we must squarely decide is whether there was legally sufficient evidence — representative, direct, circumstantial, in-person, by deposition, or otherwise — to produce a reliable and just verdict. There was. VI. WILLFULNESS AND LIQUIDATED DAMAGES Family Dollar also appeals (1) the sufficiency of the evidence to support the jury’s finding that Family Dollar willfully violated the FLSA, a decision that extended the statute of limitations from two to three years, and (2) the court’s reliance on the jury’s willfulness finding in determining that Family Dollar did not act in good faith, a decision that triggered the liquidated damages award. A. Willful Violation The statute of limitations for a claim seeking unpaid overtime wages under the FLSA is generally two years. 29 U.S.C. § 255(a). But if the claim is one “arising out of a willful violation,” the statute of limitations is extended to three years. Id. “To establish that the violation of the [FLSA] was willful in order to extend the limitations period, the employee must prove by a preponderance of the evidence that his employer either knew that its conduct was prohibited by the statute or showed reckless disregard about whether it was.” Alvarez Perez, 515 F.3d at 1162-63 (citing McLaughlin, 486 U.S. at 133, 108 S.Ct. at 1681). Federal regulations define “reckless disregard” as the “ ‘failure to make adequate inquiry into whether conduct is in compliance with the [FLSA].’ ” Id. at 1163 (quoting 5 C.F.R. § 551.104). Family Dollar raises several challenges to the jury’s willfulness finding. All fail. First, the evidence, detailed above, was legally sufficient to support the jury’s finding that Family Dollar’s FLSA violations were willful. For example, the Plaintiffs presented testimony from Family Dollar executives that it never studied whether the store managers were exempt executives. Executives also testified that Family Dollar’s company-wide policy was that store managers were exempt from FLSA overtime requirements, but they had no idea who made that policy. Further, given the evidence at trial, the jury reasonably could have found that Family *1281 Dollar executives knew that store managers spent most of their time performing manual, not managerial, tasks, that corporate manuals micro-managed store managers’ performance of those tasks, that the 380 district managers closely supervised their store managers, and that store managers had little discretion or freedom from supervision. Second, we reject Family Dollar’s suggestion that the complex and fact-intensive nature of the executive exemption inquiry means that, as a matter of law, the FLSA violations were not willful. Such a rationale would effectively preclude a willfulness finding in cases involving an executive exemption affirmative defense. While the jury could have well considered that factor in its willfulness determination, complexity alone does not preclude a willfulness finding. Third, we reject Family Dollar’s argument that the court’s decision to grant judgment as a matter of law for 163 Plaintiffs somehow biased the jury in its willfulness determination. This speculation has no support in the record. One could just as easily speculate in the other direction&emdash; the judgment for the 163 Plaintiffs meant the court thought the other 1,261 Plaintiffs failed to prove their case. Furthermore, the record shows the court instructed the jury that it still had to decide whether Family Dollar acted willfully and to determine damages for the 163 Plaintiffs. The court also instructed the jury to determine whether Family Dollar met its burden of proof on the executive exemption for the remaining Plaintiffs and, if so, to determine willfulness and damages. Finally, Family Dollar has not shown the district court abused its discretion 75 in excluding various exhibits. 76 Based on our review of the record, the district court carefully considered whether to admit these exhibits and made specific findings and rulings under Federal Rule of Evidence 403. During the first trial in 2005, the district court found that the vast bulk of this evidence was irrelevant. The district court also made a Rule 403 determination that even if there was some probative value in the evidence (although it emphasized that there was not), any such value was substantially outweighed by the other Rule 403 considerations. 77 In the second trial, the district court relied on these factual determinations and evidentia-ry rulings from the first trial. Family Dollar’s brief cites and describes the relevant exhibits. However, it fails to discuss Rule 403, to engage the court’s reason for excluding these exhibits, or to explain why the court’s particular findings and rulings were an abuse of its discretion. See Flanigan’s Enters., Inc. v. Fulton County, 242 F.3d 976, 987 n. 16 (11th Cir.2001) (stating that argument was waived because the appellants “failfed] to *1282 elaborate or provide any citation of authority in support of’ the argument in their brief). Family Dollar simply asserts that the court “excluded Family Dollar’s evidence from the jury.” In addition, Family Dollar fails to respond to the Plaintiffs’ arguments that many of these exhibits were unauthenticated, and that no evidence showed that Family Dollar’s decision-makers ever saw, or relied on, any of the documents when deciding to exempt the store managers. Viewing the record as a whole and bearing in mind the court’s wide discretion in evidentiary rulings, Family Dollar has not carried its burden to show that the court abused its discretion or that the jury’s willfulness determination was based on reversible error. B. Good Faith and Liquidated Damages When the jury finds an employer has violated the FLSA and assesses compensatory damages, the district court generally must add an award of liquidated damages in an equal amount. 29 U.S.C. § 216(b) (“Any employer who violates the provisions of ... section 207 of this title shall be liable to the employee or employees affected in the amount of ... their unpaid overtime compensation ... and in an additional equal amount as liquidated damages.”); Alvarez Perez, 515 F.3d at 1163. However, the district court has discretion to reduce or deny liquidated damages “‘if the employer shows to the satisfaction of the court that the act or omission giving rise to such action was in good faith and that he had reasonable grounds for believing that his act or omission was not a violation of the [FLSA].’ ” 78 Alvarez Perez, 515 F.3d at 1163 (quoting 29 U.S.C. § 260). “[T]he judge and jury answer what is essentially the same question for two different purposes. The willfulness or good faith question is answered first by the jury to determine the period of limitations and then, if there is a verdict for the employee, again by the judge to determine whether to award liquidated damages.” Id. at 1162. Here, the district court determined that Family Dollar failed to meet its burden of proving good faith on the liquidated damages issue because the jury already had found willfulness on the statute of limitations issue. Family Dollar argues that the district court erred because judges have the discretion to decide good faith regardless of the jury’s willfulness finding. While this was an open question in our circuit in 2006 — the time of the second trial — our subsequent decision in Alvarez Perez forecloses Family Dollar’s argument. In Alvarez Perez, we concluded that “in an FLSA case a jury’s finding in deciding the limitations period question that the employer acted willfully precludes the court from finding that the employer acted in good faith when it decides the liquidated damages question.” 79 515 F.3d at 1166. Thus, under Alvarez Perez, the district *1283 court did not err in concluding that the jury’s finding of willfulness precluded a good faith finding. Id. at 1165-66. Accordingly, the district court did not err in awarding Plaintiffs $17,788,029.32 in liquidated damages, representing an amount equal to the back pay award. 80 VII. JURY INSTRUCTIONS Family Dollar’s last claim involves the district court’s jury instructions. We start with our standard of review. Although “[w]e review jury instructions de novo to determine whether they misstate the law or mislead the jury to the prejudice of the party who objects to them,” United States v. Campa, 529 F.3d 980, 992 (11th Cir.2008) (citing United States v. Grigsby, 111 F.3d 806, 814 (11th Cir.1997)), the standard is “deferential,” Bateman v. Mnemonics, Inc., 79 F.3d 1532, 1543 (11th Cir.1996). As long as the instructions accurately reflect the law, the district court is afforded “wide discretion as to the style and wording employed in the instructions.” Bateman, 79 F.3d at 1543 (quotation marks omitted); Campa, 529 F.3d at 992; Bogle v. McClure, 332 F.3d 1347, 1356 (11th Cir.2003). Our practice is not to nitpick the instructions for minor defects. “[I]f the jury charge as a whole correctly instructs the jury, even if it is technically imperfect, no reversible error has been committed.” Bateman, 79 F.3d at 1543. “[W]e examine the challenged instructions as part of the entire charge, in view of the allegations of the complaint, the evidence presented, and the arguments of counsel, to determine whether the jury was misled and whether the jury understood the issues.” Iervolino v. Delta Air Lines, Inc., 796 F.2d 1408, 1413 (11th Cir.1986) (citation and quotation marks omitted). “When the instructions, taken together, properly express the law applicable to the case, there is no error even though an isolated clause may be inaccurate, ambiguous, incomplete or otherwise subject to criticism.” Somer v. Johnson, 704 F.2d 1473, 1477-78 (11th Cir.1983) (quoting Johnson v. Bryant, 671 F.2d 1276, 1280 (11th Cir.1982)); Verbraeken v. Westinghouse Elec. Corp., 881 F.2d 1041, 1050 (11th Cir.1989). We reverse where we are “left with a substantial and ineradicable doubt as to whether the jury was properly guided in its deliberations.” Somer, 704 F.2d at 1478 (quoting Miller v. Universal City Studios, Inc., 650 F.2d 1365,1372 (5th Cir.1981)); Johnson, 671 F.2d at 1280. After review of the jury charge as a whole and counsels’ entire closing arguments, we are convinced that the jury properly understood the issues and applicable law. Family Dollar has shown no reversible error in the charge. First, Family Dollar challenges the district court’s jury instruction on willfulness. However, the district court used the Eleventh Circuit Pattern Jury Instruction 1.7.1 that in order to prove willfulness, Plaintiffs must establish that Family Dollar knew, or showed reckless disregard for, the fact that its conduct was forbidden by the FLSA. The charge is consistent with our case law outlined earlier. See Alvarez Perez, 515 F.3d at 1162-63. The district court’s decision not to elaborate further was within its discretion. *1284 Second, Family Dollar challenges a part of the executive exemption charge. Family Dollar cannot and does not quarrel with the vast majority of the charge. That is because the court issued lengthy and comprehensive instructions on the executive exemption issue that fully captured the applicable regulations and factors. For example, the court instructed the jury that Family Dollar had to prove that: Plaintiffs made over $250 a week, their primary duties were management, and they customarily and regularly directed the work of two or more employees or their equivalent. With respect to the primary duty element, the court stated that primary duty means the main, major, or most important duty performed by the employee, and that in determining whether an employee’s duties are his or her primary duty, the jury should consider all of the facts surrounding the employment. The court also instructed that the “rule of thumb” is that primary duty means that the major part, or more than 50% of the employee’s time, was spent in performing executive duties. But it clarified that time alone is not the only factor in determining whether the employee’s duties were primarily managerial and that an employee’s primary duty may be executive even if the employee spends less than half of his or her time'in such work. In addition, the court told the jury to consider these primary duty factors: (1) the relative importance of executive duties compared with non-executive duties, (2) the frequency with which the store managers exercised discretionary powers, (3) their relative freedom from supervision, and (4) the relationship between store managers’ salaries and wages paid to other employees for nonexempt work. 81 Importantly for this case, the court also issued a concurrent duties instruction stating: “An executive employee may sometimes perform non-exempt or non-managerial duties concurrent with his executive duties, so long as the non-exempt duties are not his primary duties.” 82 Family Dollar requested this charge. In its closing argument, Family Dollar then used the concurrent duties charge to stress that store managers who performed nonexempt tasks (stocking shelves, cleaning the store, etc.) alongside their subordinates do not lose their exempt status since the store managers’ primary duty was to manage the store. Family Dollar’s principal complaint on appeal is that the court should not have added this part: “A working or supervising foreman works alongside his or her subordinates performing the same kind of work as the subordinates, and carrying out supervisory functions” and “are not executives within the meaning of the law.” Family Dollar correctly points out that there is a regulation entitled “working foremen,” see 29 C.F.R. § 541.115 (2003), and that its stated purpose is to clarify one aspect of the long test. 83 But Family Dollar argues that this means the court’s working or supervising foreman instruction does not apply in a short test case. *1285 And, therefore, the court should not have used the instruction at all. See Hays v. Pauls Valley, 74 F.3d 1002, 1008 (10th Cir.1996) (noting that the “working foreman” provision “does not apply to the short test”). But see Shockley v. City of Newport News, 997 F.2d 18, 26 (4th Cir. 1993) (noting that even when the short test applies, “[t]he ‘working foreman’ concept is useful because it helps to distinguish between a manager of a recognized subdivision and a mere supervisor of subordinate employees”). Plaintiffs respond that the short test-long test distinction was abolished in 2004, and that the concurrent duties regulation references a “working supervisor.” See 29 C.F.R. § 541.106(c) (2006) (contrasting role of working supervisor to exempt executive). Further, Plaintiffs argue that even if the working foreman regulation does not technically apply, the working or supervising foreman concept still provides a helpful example of how to apply the exemption defense. Plaintiffs point out that even we have referenced the concept of a working foreman to express ideas embodied in the short test. See Brock, 835 F.2d at 826. Although the parties hotly dispute whether the working foreman regulation applies, there is no need to resolve that thorny question in order to determine whether this particular jury charge was error here. In order for Family Dollar to prevail, it must do more than show that the working or supervising foreman charge was technically inapplicable. It must demonstrate that the charge, combined with all of the other instructions on the executive exemption issue, undermined the jury’s ability to correctly understand the applicable law and resolve the executive exemption issue. In the vernacular of our precedents, it must either mislead the jury or leave us with a “substantial and ineradicable doubt” as to whether the court properly instructed the jury. Family Dollar has not made this showing. Family Dollar overstates the significance of the working or supervising foreman charge. The charge relates to only a narrow slice of the executive exemption issue — how to deal with supervisors who work alongside of, and do the same work as, other employees. At root, it is designed to illustrate that an employee who performs the same nonexempt, manual tasks as his co-workers is not an executive even though he is technically a supervisor. Even assuming that concept is inapplicable here — a proposition that we by no means concede — the concurrent duties instruction cures any potential confusion from the working or supervising foreman charge. The concurrent duties instruction makes clear that one can still be an exempt executive even though he performs a plethora of nonexempt duties at the same time as exempt duties. Indeed, the court also accurately told the jury that a store manager may be exempt even though he spends more of his time on nonexempt duties provided that his primary duty is management. After reviewing the charge as a whole and the closing arguments, we are convinced that the jury was fully and accurately advised that an exempt executive may do exempt and nonexempt work concurrently so long as his primary duty is managerial. Although it may have been better for the court to have avoided using the working foreman charge, Family Dollar has shown no reversible error here. VIII. CONCLUSION For all of the above reasons, we affirm the district court’s judgments. AFFIRMED. Footnotes [1] Family Dollar opens around 500 stores a year. Family Dollar had 2,900 stores in 1999, 4,545 stores in 2003, 5,700 stores in 2005, and has over 6,000 stores now. Here, we recite the remaining facts in this case based on Family Dollar’s organization during the time frame relevant to this case, which is 1999 to 2005. [2] The FLSA provides that its protections "shall not apply with respect to ... any employee employed in a bona fide executive ... capacity." 29 U.S.C. § 213(a)(1). [3] In this opinion, our reference to a "Rule,” unless otherwise stated, is to the Federal Rules of Civil Procedure. [4] Although the court permitted Family Dollar to send interrogatories to the remaining opt-in Plaintiffs, it does not appear Family Dollar did so (at least from the record before us). [6] Broome testified: Q. Now, Family Dollar has not done a study about the duties of store managers in relation to whether they should be exempt or non-exempt for overtime purposes, has it? A. Not that — I don’t know, not that I'm aware of. Q. You’re the official company spokesman, as you told us; correct? Now, Family Dollar has simply said all the store managers are exempt, every one of them; right? A. Yes, Sir. Q. You said every store’s manager is considered exempt. And when you said “every,” you meant regardless of store size, number of employees, whether it's rural or urban, no matter what its profits, no matter what anything; you just said they’re all exempt, didn’t you? A. All of our store managers are salaried, yes, sir, in every store. Q. Doesn’t matter how many employees are in the store, they're all just exempt? A. All the stores of Family Dollar, yes, sir. Heskett testified: Q. Does a store’s location, whether it’s rural or urban affect whether a store manager is exempt from overtime? A. Not than I’m aware of. Q. Does the store’s size affect whether a store manager is exempt from overtime? A. Not than I’m aware of. *1246 Q. Does staff size, the number of employees supervised, does that affect whether the store manager gets paid overtime? A. Not than I’m aware of. Q. Do you know anything about stores that varies from store to store that affects whether the store managers get paid overtime? A. None that I can think of. Q. In fact the company has made a determination that all store managers for all stores will not get paid overtime at overtime rates, hasn't it? A. Our managers are salaried and have been as far as I know for a long time. [7] In May 2004, the court denied Family Dollar’s motions for summary judgment, which were based on the statute of limitations, the executive exemption defense, and judicial es-toppel. Family Dollar filed a number of motions for reconsideration, and in July 2004, the court granted Family Dollar summary judgment on the two-year statute of limitations as to 54 Plaintiffs and on the three-year statute of limitations as to eight Plaintiffs. This appeal does not involve those rulings. [8] Family Dollar repeatedly moved for judgment as a matter of law, including at the close of the evidence and after the jury’s verdict. We review a district court's denial of a motion for judgment as a matter of law de novo. Goldsmith v. Bagby Elevator Co., Inc., 513 F.3d 1261, 1275 (11th Cir.2008); Millennium Partners, L.P. v. Colmar Storage, LLC, 494 F.3d 1293, 1299-1300 (11th Cir.2007). “The question before the district court regarding a motion for judgment as a matter of law remains whether the evidence is 'legally suffi *1248 cient to find for the party on that issue/ Fed.R.Civ.P. 50(a)(1), regardless of whether the district court's analysis is undertaken before or after submitting the case to the jury.” Chaney v. City of Orlando, 483 F.3d 1221, 1227 (11th Cir.2007). “‘In considering the sufficiency of the evidence that supports the jury's verdict, we review the evidence in the light most favorable to, and with all reasonable inferences in favor of, the nonmoving party.’ If reasonable and fair-minded persons in the exercise of impartial judgment might reach different conclusions based on the evidence presented, the motion should be denied.” Millennium Partners, L.P., 494 F.3d at 1299-1300 (quoting Montgomery v. Noga, 168 F.3d 1282, 1289 (11th Cir.1999)). "We will reverse only if the facts and inferences point overwhelmingly in favor of one party, such that reasonable people could not arrive at a contrary verdict.” Goldsmith, 513 F.3d at 1275 (internal quotation marks omitted). [9] As provided in the manual, Family Dollar's computer system replenishes "basic merchandise” items. A register informs the system of every sales transaction at the store. Based on prior selling history and the current inventory level, the system determines the amount of goods needed to maintain the "maximum on hand allowance.” In addition, each store orders merchandise on a preassigned order day, which is delivered on a preassigned "truck day.” [10] For example, the top row must include, from left to right, Monthly Promotional Programs, Ads and Ad Planners, Pending Schematic and MSI Label Changes, Current Store Order and Stop Shipments, DM Notes and Latest DM Audit, Messages and EDLP Notes, and a Housekeeping Checklist. [11] Drawer One contains the petty cash box, bank deposit bags, register keys, extra keys, stamps, and an extra till bag. Drawer Two houses items such as unauthorized price change reports and area change reports. Drawer Three holds the open direct summary file, packing slips, credit memos, direct ship *1251 ment notifications, and processed billing summaries. Drawer Four has messages, payroll, personnel, drug screening, resignation verification, employee evaluations, warning forms, blank applications, completed applications, and a variety of other forms. [12] "Q. But they have to get district manager approval if they go outside the budget, and are subject to discipline if they don’t? A. They could be subject to discipline, but there's many times that they don’t get approval to go outside the budget. Q. But there’s also been times, as we see in these emails, where they're threatened with their job if they go one penny over their budget that's been given to them by their district manager; correct? A. I’m sure you have a document that shows that, yes.” [13] "Q. And the budget limits how much payroll they have; in fact, it's just a payroll budget, isn’t it? A. The budget's based on sales and store hours and then other factors. Q. Those are the elements that go into the budget. But the budget itself is just covering payroll, nothing but payroll; correct? A. It’s covering hours, associate hours. Q. It tells how many associate hours that managers are allowed to have? A. Yes. Q. If that’s not enough, he has to do it himself, work whatever hours it takes; right? A. Yeah. The job has to get done.” [14] Plaintiffs’ counsel confronted Barkus with his deposition testimony: Q. But you’ve known — we took your deposition in 2002 — you told us then that that’s the way the system works; the budget staffs the store at just a certain level, and the store manager has to make up all the other hours? A. Can you point that out to me? Q. Yes. Turn to Page 322 of your deposition. And, again, it’s shifted a little bit. It’s 324, line 11. “Question; In many instances, the store managers will assume the duties of those associates so that their budget stays in line, or they come in under budget; correct? Answer: Yes.” A. That's correct. Q. And, in fact, the people under you, your district managers, they have sent emails out saying to the store managers, "I'm cutting your budget.” It means, I'm taking away the hours of your subordinates; and you work the hours, 60, 70, 80, hours, whatever, haven’t they? A. I haven’t seen it, but I’m sure you're referring to something. Q. Yes. Exhibits 18 and 19, 25 through 29, 32, 33, 45, 154, 168; you go through the book, you see it over and over. I’m taking your budgets away. You work the hours. [15] "Q. Do you recall the ‘Essential Job Functions of the Assistant Store Manager’? A. Yes, I do .... Q. And the assistant store manager job description, called Essential Job Functions, is Exhibit 7. It shows that the assistant store manager, his regularly assigned essential duties are the same ones as for the store manager; virtually word for word; correct? A. Yes.” Other evidence showed that hourly assistant managers performed the same tasks as the store manager. [16] "Q. Now, Family Dollar has not done a study about the duties of store managers in relation to whether they should be exempt or non-exempt for overtime purposes, has it? A. Not that — I don't know, not that I’m aware of.” [17] “Q. If a store manager, though, goes over budget, he spends more than what he's been given by his district manager, he can be disciplined; right? A. Could be, yes .... Q. But you remember when I showed you store managers who were — they were threatened to be fired for going over budget; you remember that, don’t you? A. There have been e-mails written stating that, yes, sir. Q. And those emails came from district managers to store managers telling them they'd be disciplined or fired if they spent one dime, one penny, *1254 over the budget that that district manager had given them; right? A. I don’t remember the specific words, but, yes. They said that you could be disciplined up to and including termination for doing something.” [18] ”Q. And it always shows 52 hours for the store manager; correct? A. To the best of my knowledge, yes. Q. Because the system is built — that is, the staff scheduler system and the store payroll budget is built on the 52-hour week assumption for the last several years, for store managers; right? A. The staff scheduler is built with 52 hours as a base for the manager, the manager's salary. So the number of hours they could work 40 and still receive their salary, or they could work 60 and still receive their salary. But it's built using that as part of the coverage.” [19] Some district managers showed on the staff scheduler the days and hours that assistant managers and sales associates worked. But others allowed their store managers to plug in who worked when so long as they did not exceed the total allotted labor hours for the week. [20] As Broome testified; "Q. Can I move coverage from slower days, (i.e., non-truck weekdays) to Fridays, Saturdays or Sundays when we are busy? A. No. Store's schedules are designed to give coverage that are based on store volumes, store hours. Q. So this is telling store managers they can't schedule people the way they want to schedule them; correct? A. It’s basically telling them they should not move coverage from slow days to the weekend.” [21] A portion of Family Dollar’s Professional Development Training Reference Book describes the “Family Dollar Store Team” in order of seniority as District Manager, Store Manager, Assistant Manager, and Associates. [22] District managers visit each store at least once a month and some visit twice a month or more. The job description for a district manager states: "Visit stores, continuously developing plans to improve performance. One store per day is to be worked Monday through Thursday, with multiple stores on Friday and Saturday if desired unless otherwise directed.” [23] Assistant managers work full-time, while sales associates work either part- or full-time. [24] According to the manual, district managers monitor the payroll budget by requiring every store manager to key payroll hours into the Payroll Daily Form. Store managers must enter payroll hours at the end of the day for employees that have clocked out for the day or "first thing in the morning for the employees who worked the previous day.” Every Wednesday, store managers are required to print out and transmit the Payroll Daily Form. At the end of each week, store managers must record weekly net sales (i.e., the figure from the "Beat Yesterday Book”) and actual payroll dollars spent, and send this information to their district managers. [25] The manual requires all Family Dollar stores to be open seven days a week for the entire year, except for Christmas Day. Corporate headquarters sets the hours — nine a.m. to seven or eight p.m. on Monday through Saturday, and noon to six p.m. on Sunday. Each store is open 72 hours a week. Only headquarters or a district manager can change a store's hours. The manual states that "there are only two sets of keys to a Family Dollar Store.” One set is for the store manager, who "should maintain possession of his or her keys at all times,” and the other set is held by the assistant manager or the floor supervisor. The manual allows for a third key to be given to another employee for emergency purposes, *1256 but that employee must be recommended by the store manager and approved by the district manager. [26] Broome testified about hiring: "Q. In fact, if a store manager were to attempt to hire his own assistant without the district manager interviewing and selecting the person, they could be disciplined and fired, couldn’t they? A. If they had not been given that authority by the district manager, they could be. Yes, sir. Q. I'm talking about the company policy is, that if a store manager tries to hire his own assistant instead of having the district manager do the interview and doing the hiring, they are subject to discipline, aren’t they? A. They could be, yes, sir.” [27] Barkus testified: Q. And you don’t know of any training or instructions that have been given to store managers which tells them that they have the power to discharge a store employee, do you? A. I don’t know of any, no. Q. And you don’t know of any instructions or anything the store managers have been told that says that they can suspend an employee in their store on their own either, do you? A. I don't know of any, no. Q. And, in fact, you issued an email in February 2003, which told the store managers that if they tried to hire their own assistants on their own, they would be subject to being fired, terminated, didn’t you? A. Yes, I did. Q. And you recall no point in time during your seven years at Family Dollar that that wasn’t the policy, do you? A. No, I don’t. [28] In addition, "Strand 4” of Family Dollar’s Professional Development Training Reference Book provides that, "You must contact your District Managers or Regional Vice President before conducting a Discharge.” [29] Some district managers allow their store managers to ignore corporate policy and hire candidates for hourly sales associates positions without obtaining district manager approval. The manual states the district manager usually meets the associate on her next store visit. [30] The manual indicated that district managers must approve hourly associate raises. “The District Manager will review wage increase recommendations with the Regional Vice President. Store Managers will be informed whether or not the recommendations are granted.” [31] Plaintiffs' Trial Exhibit 294 (referencing various portions of "Strand 4”) states that: The Retention Review allows the Store Manager and District Manager to conduct a 90-day performance appraisal, provide feedback and coaching for Store Associates and to advise them of their eligibility for certain employee benefits .... DM will schedule in-store visits with all associates who will soon "hit” their 90-day milestone .... Store Manager and District Manager meet with Associate and review information contained in Retention Review. [32] These are the average salaries for all Plaintiffs (using their Exhibit 278) and Family Dollar’s average salaries for all store managers nationwide (using its Exhibit 2324): All Plaintiffs’ Salaries (526) 1999: $523 (882) 2000: $542 (940) 2001: $555 Nationwide Salaries 1999: $553 2000: $573 2001: $596 [33] Average hourly salary for store managers: Store Manager’s Salary -j- 60 hours: 1999: $ 8.72 2000: $ 9.03 Store Manager’s Salary 70 hours: 1999: $ 7.47 2000: $ 7.74 Assistant Manager’s Hourly Wage 1999: $7.11 2000: $7.19 (684) 2002 $581 2002 $601 (249) 2003 $623 2003 $606 (123) 2004 $668 2004 (86) 2005 $706 2005 $651 The two sets of figures do not differ materially. For example, there is only a 5% difference between the 1999 salaries. Plaintiffs’ figures (for 2003 to 2005) benefit Family Dollar by suggesting that Plaintiffs made more per hour than Family Dollar attempted to show. Store managers may earn a bonus. Although Family Dollar’s brief cites testimony by a few store managers showing that the average bonus size was $1,800, it is difficult to factor the bonus into the salary spread between store managers and assistant managers without an indication of how often Family Dollar paid such bonuses and to what percentage of store managers. In any event, the bonus does not substantially increase the range of pay. For example, even if every store manager earned a $1,800 bonus every year, that translates into $34.61 more per week. Assuming a 60-hour workweek, that equals an extra $.58 more an hour. Assuming a 70-hour week, it amounts to $.49 more an hour. These small amounts do not alter the analysis of whether store managers earn "significantly" more than assistant managers. *1258 2001: $ 9.25 2001 $ 7.93 2001: $7.43 2002: $ 9.68 2002 $ 8.30 2002: $7.65 2003: $10.38 2003 $ 8.90 2003: $7.79 2004: $11.13 2004 $ 9.54 2004: $7.88 2005: $11.77 2005 $10.09 2005: $8.18 [34] This $1,575,932.12 to 163 Plaintiffs and $17,516,071.27 to the remaining Plaintiffs totaled $19,092,003.39. Although Plaintiffs requested a much larger amount of back pay based on a rate of pay using either a 40-hour or 48-hour workweek, the $19,092,003.39 sum represented Family Dollar's own calculation of overtime back pay based on Defendant’s Trial Exhibit 1959, and reflected a rate of pay based on a salary intended to compensate for a 52-hour workweek. In other words, Family Dollar’s chart divided the store managers' base salary by 52 hours (not 40 or 48 hours) to determine the hourly rate of pay for Family Dollar’s overtime pay calculations. Thus, the jury’s use of Family Dollar’s calculations substantially reduced the verdict amount, and there is no cross-appeal on that issue. After the verdict, the district court twice adjusted the back pay amount for certain Plaintiffs for various reasons (such as bankruptcy, judicial estoppel, and standing), and ultimately entered a March 31, 2006 judgment for back pay of $16,623,989.32, and increased the amount by $1,164,040.42 on April 6, 2007, for a total back pay judgment of $17,788,029.74. [35] In Albritton, we slated: "[Section] 216(b) requires that would-be plaintiffs affirmatively opt in, and that they do so in writing, and that the writing be filed in court before they can be included in the lawsuit.” 508 F.3d at 1017. The FLSA “does not indicate that opt-in plaintiffs have a lesser status than named plaintiffs insofar as additional claims are concerned. To the contrary, by referring to them as 'party plaintiff[s]’ Congress indicated that opt-in plaintiffs should have the same status in relation to the claims of the lawsuit as do the named plaintiffs.” Prickett v. DeKalb County, 349 F.3d 1294, 1297 (11th Cir.2003). [36] Unlike opt-in collective actions under § 216(b), a district court's decision to certify a Rule 23(b) class binds all the class members. See Cameron-Grant, 347 F.3d at 1249 (detailing the wide range of differences between § 216(b) collective action and Rule 23 class action); see also Grayson v. K Mart Corp., 79 F.3d 1086, 1096 n. 12 (11th Cir. 1996) ("[I]t is clear that the requirements for pursuing a § 216(b) class action are independent of, and unrelated to, the requirements for class action under Rule 23 of the Federal Rules of Civil Procedure.”). [37] Although Hipp involved a collective action brought under the Age Discrimination and Employment Act of 1967, 29 U.S.C. §§ 621 et seq., that statute incorporates the FLSA's collective action mechanism, see 29 U.S.C. § 626(b). Therefore, “Hipp ... applies in both contexts.” Albritton, 508 F.3d at 1014 n. 1; Cameron-Grant, 347 F.3d at 1243 n. 2. [38] It appears that we are not the only ones. See Thiessen v. Gen. Elec. Capital Corp., 267 F.3d 1095, 1105 (10th Cir.2001) (referring to Eleventh Circuit’s "ad hoc” approach as "[a]rguably ... the best of the three approaches” taken by various circuits, refusing to endorse any particular approach, and reviewing for "abuse of discretion”); Mooney v. Aramco Servs. Co., 54 F.3d 1207, 1213-16 (5th Cir.1995) (describing it as "remarkable” that a number of courts have failed to formally define "similarly situated,” referring to such approaches as "ad hoc,” but refusing to adopt any formal definition or set of factors), overruled on other grounds by Desert Palace, Inc. v. Costa, 539 U.S. 90, 123 S.Ct. 2148, 156 L.Ed.2d 84 (2003). [39] "In Grayson, we interpreted the FLSA’s collective action provision to require plaintiffs alleging age discrimination to 'demonstrat[e] a reasonable basis for their claim of class-wide discrimination ... by making substantial allegations of class-wide discrimination, that is, detailed allegations supported by affidavits which successfully engage defendants’ affidavits to the contrary.' ” Anderson, 488 F.3d at 952. Likewise, in Haynes, we stated: "As a preliminary matter, it is not disputed that plaintiffs have the burden of demonstrating a reasonable basis for crediting their assertions that aggrieved individuals existed in the broad class that they proposed.” Haynes, 696 F.2d at 887. [40] District courts following the two-step Hipp approach should treat the initial decision to certify and the decision to notify potential collective action members as synonymous. Indeed, Hipp collapses the two questions by describing the first stage of certification as "the notice stage.” Hipp, 252 F.3d at 1218. Our circuit precedent on notice is, for the purpose of a Hipp analysis, the same as our law on the question of conditional, and initial, certification. [41] In some cases, the district court's first-stage certification analysis is properly based on plaintiffs' pleadings and affidavits, Anderson, 488 F.3d at 952-53, or on affidavits and deposition transcripts, Cameron-Grant v. Maxim, 347 F.3d 1240, 1244 (11th Cir.2003); Grayson, 79 F.3d at 1097. Sometimes only limited information is available at the notice stage and detailed pleadings and affidavits may suffice. We do not hold that the amount of discovery that preceded the first-stage certification here is required but only that the district court had an ample "reasonable basis” to support its 2002 certification order. [42] In its 2002 certification order, the district court found that Plaintiffs could not hire, discipline, or terminate employees without district manager approval. In its 2005 order denying decertification, the court narrowed this to assistant managers, finding store managers could not hire, discipline, or terminate assistant managers without district manager approval. As discussed earlier in section II, store managers were generally not authorized to hire or fire assistant managers without district manager approval, which supports the district court’s 2005 finding. The evidence shows, however, that store managers could give verbal and written discipline warnings to all employees, including assistant managers. So although the district court erred in this finding as to discipline, the store managers nonetheless remained similarly situated as to discipline of assistant managers and other employees. [43] Although the Grayson and Hipp plaintiffs worked in several states, this Court concluded that they met the similarly situated requirement. Grayson, 79 F.3d at 1091; Hipp, 252 F.3d at 1219. [44] Family Dollar relies heavily, both in the district court and on appeal, on charts that summarize the store managers’ deposition testimony. Family Dollar argues the charts illustrate that store managers' duties differed substantially. Plaintiffs also submitted charts that summarize the similarity in their duties. These charts were not used at trial — only at the decertification stage. The parties' respective charts not only conflict as to what the deposition testimony *1264 shows, but also fail to paint a full picture. For example, according to Family Dollar’s chart, Mary Kimsey had the "authority to hire" (depicted by a "Y” in the "Authority to Hire” column). But Kimsey's deposition shows that she lacked the power to hire assistant managers (although she did have the power to hire sales clerks). The chart suggests that she had the power to discipline subordinates. But according to her deposition, her ability to discipline was contingent on district manager approval. Where Family Dollar's chart indicates that Kimsey "set or adjusted] the hours of employees,” she actually testified that she had some flexibility to adjust hours, but that she was heavily constrained by the corporate planner and the district manager. Furthermore, Family Dollar's charts do not include all of the factors, discussed later, that courts examine when making an executive exemption inquiry. Family Dollar’s charts mostly addressed a store manager's authority to hire, fire, interview, train, and discipline other employees. But they did not address many other aspects of a store manager's day-to-day duties. In any event, the district court was free to assess the quality of the charts, and there was ample evidence to support the court's fact-findings that the Plaintiffs were similarly situated. [45] We do not suggest that a collective action needs to have 14 significant job factors in common in order to be, or stay, certified or that less commonality would not have survived a decertification motion. Rather, we highlight the number of similarities only to show that the decertification decision is not close here. [46] Just because a business classifies all employees in a particular job category as exempt does not mean that those employees are necessarily "similarly situated” for purposes of a 29 U.S.C. § 216(b) collective action. Rather, it is necessary to review the actual job duties of those in that job category to determine whether they are similarly situated and whether the exemption defense can be collectively litigated. Here, Family Dollar argues the store's size, sales volume, and location cause store managers’ job duties to vary and preclude a collective trial. The facts — that Family Dollar never examined how store managers spent their time and that none of those factors had anything to do with Family Dollar’s decision to exempt all store managers from overtime pay-counter Family Dollar's argument in that regard. [47] On appeal, Family Dollar does not contend it was denied a fair opportunity to present its statute of limitations defense. Family Dollar produced detailed records of each of the 1,424 Plaintiffs’ employment dates. Since the date on which each Plaintiff joined this lawsuit is a matter of public record, the parties submitted charts showing the amount of back pay owed for each plaintiff for both a two-year and a three-year time period. Thus, once the jury decided the executive exemption issue in Plaintiffs' favor, the jury had the necessary evidence in chart form to limit Plaintiffs’ back pay to either two or three years. This underscores how collective actions about overtime pay can be readily and fairly managed. Plus, this is not a case of borderline similarity among the Plaintiffs, but one of heightened similarity, which significantly adds to our fairness conclusion. [48] The old regulations had a long test and a short test. The long test applied to those who made not less than $155 a week. 29 C.F.R. § 541.1(a)-(f) (2003). The short test, on the other hand, applied to those making not less than $250 a week. 29 C.F.R. § 541.1(f) (2003). The entire test was contained in section (0, and it was therefore "short" in comparison to the longer test contained in § 541.1 (a)-(£) (2003). See Atlanta Prof'l Firefighters Union, Local 134 v. City of Atlanta, 920 F.2d 800, 804-05 (11th Cir.1991). The current regulations (effective August 23, 2004) abolished the distinction between the long and short tests. See Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 69 Fed.Reg. 22,122, 22,122-25 (Apr. 23, 2004) (codified at 29 C.F.R. pt. 541) (explaining reasons for the August 23, 2004 change). The Plaintiffs whose claims involve conduct that occurred prior to August 23, 2004 all earned more than $250 a week. Therefore, only the short test under the old regulations and the August 23, 2004 regulations apply to this case. [49] We cite the pre-August 23, 2004 regulations with a 2003 citation and the post-August 23, 2004 ones with a 2006 citation. [50] The new regulations explain the determination of whether an employee's recommendations are given particular weight includes consideration of: whether it is part of the employee’s job duties to make such suggestions and recommendations; the frequency with which such suggestions and recommendations are made or requested; and the frequency with which the employee's suggestions and recommendations are relied upon .... An employee’s suggestions and recommendations may still be deemed to have "particular weight" even if a higher level manager's recommendation has more importance and even if the employee does not have authority to make the ultimate decision as to the employee's change in status. 29 C.F.R. § 541.105 (2006). [51] While the new regulations elevate the employee’s hiring and firing authority to a fourth, separate requirement, the old regulations likewise considered the selection of employees as a management task under the primary duty inquiry. [52] The new regulations list these management tasks: [Interviewing, selecting, and training of employees; setting and adjusting their rates of pay and hours of work; directing the work of employees; maintaining production or sales records for use in supervision or control; appraising employees' productivity and efficiency for the purpose of recommending promotions or other changes in status; handling employee complaints and grievances; disciplining employees; planning the work; determining the techniques to be used; apportioning the work among the employees; determining the type of materials, supplies, machinery, equipment or tools to be used or merchandise to be bought, stocked and sold; controlling the flow and distribution of materials or merchandise and supplies; providing for the safety and security of the employees or the property; planning and controlling the budget; and monitoring or implementing legal compliance measures. 29 C.F.R. § 541.102 (2006). [53] The new regulations list these factors to consider for the primary duty inquiry: [T]he relative importance of the exempt duties as compared with other types of duties; the amount of time spent performing exempt work; the employee’s relative freedom from direct supervision; and the relationship between the employee's salary and the wages paid to other employees for the kind of nonexempt work performed by the employee. 29 C.F.R. § 541.700(a) (2006). [54] The new regulations illustrate the concept of "concurrent duties”: For example, an assistant manager in a retail establishment may perform work such as serving customers, cooking food, stocking shelves and cleaning the establishment, but performance of such nonexempt work does not preclude the exemption if the assistant manager’s primary duty is management. An assistant manager can supervise employees and serve customers at the same time without losing the exemption. An exempt employee can also simultaneously direct the work of other employees and stock shelves. 29 C.F.R. § 541.106(b) (2006) (emphasis added). [55] The new regulations illustrate this proposition through this example: [A] relief supervisor or working supervisor whose primary duty is performing nonexempt work on the production line in a manufacturing plant does not become exempt merely because the nonexempt production line employee occasionally has some responsibility for directing the work of other nonexempt production line employees when, for example, the exempt supervisor is unavailable. Similarly, an employee whose primary duty *1269 is to work as an electrician is not an exempt executive even if the employee also directs the work of other employees on the job site, orders parts and materials for the job, and handles requests from the prime contractor. 29 C.F.R. § 541.106(c) (2006) (emphasis added). [56] Family Dollar does not argue that the old and new regulations lead to different results in the primary duty inquiry here. Thus, we need not distinguish between them in our analysis. Although Rodriguez was decided under the old regulations, 518 F.3d at 1262 n. 2, its focus on the "fact-intensive nature of the primary duty inquiry” is correct under the new regulations too, see 29 C.F.R. § 541.700(a) (2006). [57] As noted above, the new regulations omit this discretionary-power factor, but they retain the freedom-from-supervision factor. Having discretionary power is one aspect of freedom from supervision. In addition, the new regulations specify that the determination of an employee’s primary duty includes, but is not limited to, the listed factors and is to be made “with the major emphasis on the character of the employee’s job as a whole.” 29 C.F.R. § 541.700(a) (2006). Thus, the new regulations do not preclude, and are consistent with, our consideration of the frequency with which an employee exercises discretionary powers in our primary duty analysis. [58] Although Family Dollar emphasizes its store managers had management-type paperwork to do (such as bank deposits and accident and payroll reports), Family Dollar ignores that how to do each task was prescribed by the manual and district managers and store managers had little discretion, under Plaintiffs' evidence, as to those tasks. [59] The district court explicitly instructed the jury on these factors. [60] Importantly, the Sixth Circuit noted that courts cannot rely solely on whether an employee was “in charge” of the store. Thomas, 506 F.3d at 503. Rather, they must evaluate the employee's actual job duties. Id. The court did "not adopt a rule that any employee who is in charge of a store has management as her primary duty; [it] merely conclude[d] that other cases stating as much do not conflict with [Sixth Circuit] precedent.” Id. at 503 n. 4. "When a court is asked to consider whether an employee's primary duty consists of management, the proper analytical approach is to scrutinize the factors in the Secretary's regulations, not simply to determine whether the employee was ‘in charge.' " Id. [61] The Baldwin decision also rests on other materially dissimilar facts. See Baldwin, 266 F.3d at 1108-09, 1115 (stating plaintiffs "had to adhere to company policies” but describing plaintiffs as "free from daily supervision,” noting that supervisor "visited the park once or twice a month, and there was no constant oversight,” and that "in practice, the oversight was neither so rigorous nor so frequent as to undermine the undeniable fact that the Baldwins were substantially free from supervision”). Further, the Baldwin plaintiffs managed without much interference by their corporate headquarters. Id. at 1115 ("They managed the park without much participation or interference from Trailer Inns, Inc.’s Yakima headquarters.”). As noted in Rodriguez, Baldwin "illustrates the necessarily fact-intensive nature of the primary duty inquiry.” Rodriguez, 518 F.3d at 1264. [62] Family Dollar stresses that its store managers either hired and fired the hourly employees, or had district managers who frequently followed their hiring and firing recommendations. Hiring and firing authority is now a fourth, separate requirement in the executive exemption test; even if met, Family Dollar still has to establish the store managers' primary duty was management. 29 C.F.R. § 541.100(a) (2006); 29 C.F.R. § 541.1 (2003). While this is certainly a fact the jury could also consider in making that primary duty determination, the jury was required to base its decision on all the facts and was not bound to find the store managers’ primary duty was management due to this authority over hourly employees. [63] The old regulations did not define "customary and regular” in the short test for the executive exemption. However, the long test under the old regulations did define the phrase "customarily and regularly,” albeit with regard to the long test's separate requirement that an exempt executive "customarily and regularly exercises discretionary powers” in 29 C.F.R. § 541.1(d) (2003). It provided that "customarily and regularly” is "a frequency which must be greater than occasional but which, of course, may be less than constant.” 29 C.F.R. § 541.107(b) (2003). Family Dollar cites § 541.107 as a useful definition of "customary and regular.” [64] To be clear, Family Dollar also submitted other exhibits. But the others assumed that a full-time employee worked a 30-hour workweek (Exhibit 1742A) or a 35-hour workweek (Exhibit 1742B). The district court used Exhibit 1742C (based on a 40-hour workweek), which is why we focus on that exhibit. [65] The DOL Wage and Hour Division’s Field Operations Handbook similarly states that "[t]wo full-time employees or the equivalent within the meaning of Reg. 541.105(a) is generally considered to mean 80 hours of work by subordinate employees.” Field Operations Handbook, § 22c00(b). This Court has noted that the DOL's Field Operations Handbook is persuasive, even though it is not entitled to Chevron deference. Klinedinst v. Swift Invs., Inc., 260 F.3d 1251, 1255 (11th Cir.2001). [66] In fact, Plaintiffs challenged Family Dollar’s exhibits in part on the grounds that Family Dollar’s calculation of the total employee hours failed to account for whether the store managers were physically present at the store at the same time as their subordinates. In other words, even where store managers were absent, Family Dollar still counted all subordinate labor hours for the purpose of their exhibits on the customary and regular supervision issue. The district court rejected the Plaintiffs' challenge. Therefore, Family Dollar benefited from every subordinate hour worked — without regard to whether the store managers were physically present or on vacation, out sick, on leave, or just not there. [67] Indeed, at Family Dollar’s request, the district court instructed the jury that ”[c]ontinuous physical presence is not an essential requirement for supervision.” [68] To establish its prima facie case, Plaintiffs demonstrated that: (1) Family Dollar employed them; (2) Family Dollar is an enterprise engaged in interstate commerce covered by the FLSA; (3) each Plaintiff actually worked in excess of a 40-hour workweek; and (4) Family Dollar did not pay any overtime wages to them. See 29 U.S.C. § 207(a). [69] It appears that the district court required the Plaintiffs, as part of their case-in-chief, to put up their rebuttal to Family Dollar’s executive exemption defense before Family Dollar itself called any witnesses. Even so, the Plaintiffs presented more evidence as to the executive exemption in their case-in-chief than just the seven testifying Plaintiffs. And in any event, Plaintiffs established their prima facie case, and the executive exemption is Family Dollar's affirmative defense — not a part of Plaintiffs' case-in-chief. Thus, we view the evidence on the executive exemption defense as a whole. [70] MR. ST. CLAIR [for Family Dollar]: We had our extracts, Your Honor. They offered and testified the balance of those depositions. We're now talking about something completely different, depositions for which we haven't offered anything into evidence. MR. JOHNSON [for Plaintiffs]: That goes to balance. MR. ST. CLAIR: If it was just a party, it would be like I take my own deposition and put it in at trial. We can't do that. I might also add— MR. R. WIGGINS [for Plaintiffs]: They wanted to take 250 depositions, Your Hon- or, of the collective group. You allowed them to take 250. They now have offered, I’m told, twelve. And we’re offering the other, I guess, 238. THE COURT: And you're offering them for the truth of the matter asserted? MR. R. WIGGINS: Yes, as sworn testimony under the rules. Yes. MR. JOHNSON: Yes. THE COURT: And you — and you’ve not sought a stipulation as to what these witnesses would say, if called? MR. R. WIGGINS: During the earlier part of the case, Judge, we could not come to an agreement about stipulations on that. MR. ST. CLAIR: Your Honor, this is the first time that I've heard anything about them offering depositions in this case from people who were not called as witnesses. *1278 This is the very first time that issue has been raised, Your Honor. MR. R. WIGGINS: It’s not true, and it’s mainly because he just entered the case recently, Your Honor. These are on our exhibit list. These — and there’s nothing surprising about it. We’ve argued this before, it’s just not — it's no surprise to them. We’ve always said, if you offer part of 250, we're going to offer the other part. MR. ST. CLAIR: And, Your Honor, I'm told, at the pretrial conference said they will not be admitted into evidence. I wasn’t at the pretrial conference, but my partner was. THE COURT: Well, Plaintiffs’ Exhibit 297 as offered by the plaintiff is not received in evidence, but it will be a part of the record. Any other evidence from either party? [71] Although Plaintiffs’ offer of the deposition testimony was not received in evidence, the court ruled that it would “be a part of the record.’’ [72] Family Dollar’s cross-examination of Plaintiffs' witnesses did not count against its 40 hours unless its cross-examination time exceeded that of Plaintiffs’ direct-examination. There is no indication in the briefs that Family Dollar’s cross-examination reduced any of its 40 hours. [73] Congress abrogated a portion of the Mt. Clemens opinion not relevant here with the passage of the Portal-to-Portal Act, Pub.L. No. 49-52, 61 Stat. 87 (1947). See 29 U.S.C. § 251; United States v. Cook, 795 F.2d 987, 990-91 (Fed.Cir.1986). [74] Even in non-Mf. Clemens-lype cases, courts have authorized representative testimony in FLSA cases. See, e.g., Burger King I, 672 F.2d at 225 (authorizing district court to rely on representative testimony in FLSA case to prevent cumulative testimony); Dole v. Snell, 875 F.2d 802 (10th Cir.1989). [75] We review a district court’s rulings on the admissibility of evidence for abuse of discretion and will reverse only if the moving party establishes a substantial prejudicial effect. Goldsmith v. Bagby Elevator Co., Inc., 513 F.3d 1261, 1276 (11th Cir.2008). [76] Family Dollar focuses mainly on these exhibits: (1) a district court decision (Exhibit 1955) and a magistrate judge decision (Exhibit 1954), both finding that a store manager was exempt from the FLSA’s requirements; (2) correspondence to and from DOL Wage and Hour investigators (Exhibits 2249 and 2336) and state investigators (Exhibits 2356, 2357, and 2249); and (3) a report from two psychologists. [77] Federal Rule of Evidence 403 authorizes district courts to exclude evidence "if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.” Fed.R.Evid. 403. [78] "The employer bears the burden of establishing both the subjective and objective components of that good faith defense against liquidated damages.” Alvarez Perez, 515 F.3d at 1163; Dybach v. State of Fla. Dep’t of Corr., 942 F.2d 1562, 1566-67 (11th Cir.1991). [79] In Alvarez Perez, we explained this conclusion, in part, this way: [Wjhen a jury finds that a defendant's violation is willful for statute of limitations purposes, it has already factored the possibility of good faith into its examination .... Not , only would a district court impermissively be making a finding contrary to the jury’s findings, but ... to find "good faith” after a finding of "willful” violation is illogical; the two terms are now mutually exclusive. Alvarez Perez, 515 F.3d at 1165 (quotation marks omitted). [80] This Court recognized in Alvarez Perez that, in light of its holding, “evidence that an employer acted without willfulness and in good faith makes a difference at this stage only if that evidence compels judgment as a matter of law for the employer." Id. at 1167. Based on the Plaintiffs’ evidence discussed above in the analysis of the jury’s willfulness finding, we cannot say that Family Dollar's evidence compelled judgment as a matter of the law in its favor on willfulness and, in turn, on good faith. [81] See 29 C.F.R. § 541.700(a) (2006); 29 C.F.R. § 541.103 (2003). [82] The court also defined exempt and nonexempt work, and emphasized that continuous physical presence at a store is not an essential requirement for determining whether an employee supervised two or more employees. The court instructed that "job title alone” is not enough to determine whether an employee is exempt, but rather that the jury must examine “the employee’s salary and actual job duties.” The court also instructed that work that is directly and closely related to executive work is considered executive work, even if it includes routine menial tasks. [83] See supra note 48.