Vitali v. Reit Management & Research, LLC

Milkey, J. The plaintiff, Donna Vitali, worked as a bookkeeper for the defendant, Reit Management & Research, LLC (company), a property management firm. She was paid by the hour, and pursuant to both statute and company policy, she was to be paid overtime at one and one-half times the regular rate for any work done in excess of forty hours in a given week. See G. L. c. 151, § 1A. She brought the current action alleging that she accrued overtime that was not credited by the system the company had in place to keep track of employee hours. In a detailed and thoughtful decision, a Superior Court judge allowed the company’s motion for summary judgment. Because we conclude that there are material facts in dispute, we reverse. Standard of review. Our review of the allowance of a motion for summary judgment is de novo. Deutsche Bank Natl. Trust Co. *100 v. Fitchburg Capital, LLC, 471 Mass. 248, 252-253 (2015). Disputed facts are to be read in the light most favorable to the nonmoving party, in this case, Vitali. Id. at 250. “The moving party must affirmatively show that there is no real issue of fact, all doubts being resolved against the party moving for summary judgment.” Shawmut Worcester County Bank, N.A. v. Miller, 398 Mass. 273, 281 (1986) (quotation omitted). Evidence in the record is considered together with all reasonable inferences to be drawn from the record. Godfrey v. Globe Newspaper Co., 457 Mass. 113, 119 (2010). Background. 1. The nature of the dispute. Vitali was scheduled to work from nine to five, five days per week, with a paid one-hour lunch break. Both sides agree that lunch breaks do not count toward overtime. They also agree that if an employee has to work during what otherwise would be a lunch break, the employee gets no extra pay for doing so (since she or he is already being paid for that time). However, such worked lunch time can be counted toward the forty-hour overtime threshold, thus potentially indirectly increasing the employee’s over-all compensation. Vitali claims that she regularly worked during her lunch breaks even though that time was not recorded in the particular timekeeping system that the company used during the relevant period. She brought this action pursuant to G. L. c. 151, § 1A, purportedly as a class action, seeking the extra compensation that would be due if she and others similarly situated were credited for such lunch time work. 1 2. The company’s timekeeping system. On February 15, 2010, the company implemented a new electronic timekeeping system. 2 Under this system, which was known as Kronos, hourly employees were required to use their computer terminals to “punch in” when they first arrived on a given day, and to “punch out” when they left. At the center of this case is how the company, relying on Kronos, accounted for employee lunch breaks. As the company acknowledged, when Kronos was first implemented, it did not have the “functionality” to allow employees to punch out for lunch and to punch back in when they returned. The absence of *101 that feature created a potential discrepancy between the hours that an employee “clocked” using Kronos and the time they actually worked. Thus, for example, if Vitali confined her work to the scheduled nine-to-five work day and took her allotted one-hour paid lunch breaks, she would clock forty hours even though she actually worked only thirty-five hours. As a result, if Vitali performed work outside of the ordinary nine-to-five work day, the time automatically would be captured as clocked hours, but any time she spent working during lunch would not similarly be reflected. Thus, regardless of whether Vitali worked through all (or part) of lunch or took her full allotted lunch break, her hours clocked in Kronos would be the same. 3. The company’s practice in calculating overtime. In light of the discrepancy between hours worked and hours clocked, the company adopted a practice of paying overtime to hourly employees only once they clocked forty-five hours for a given week unless the employees separately reported having to work in lieu of lunch. In other words, except to the extent that hourly employees separately recorded their lunch time work, the company assumed that they took their full one-hour lunch breaks. According to Melissa Juppe, the company’s payroll supervisor, the proper protocol for recording lunch time work in Kronos was for employees to access a “drop down” menu on their computer screen through which they could then input the time code “worked hours” for the relevant amount of time. 3 In Juppe’s own words, employees “would have to log in and then once they’re on their timecard, they go to the day they didn’t take their lunch, they insert a row and the pay code column they’d do the drop down and there’s a code that says working hours, and they would record the time that they worked during their lunch.” The extent to which employees were informed of this procedure and instructed that they should use it is reserved for later discussion. 4. Vitali’s alleged lunch time work. The exigencies of the company’s property management responsibilities sometimes required employees to work beyond their scheduled hours. For those in Vitali’s position, the events that required extended work included mass lease terminations, “[mjonthly closes, quarter closes, conference calls for bad debt, [and] audits.” As noted, when hourly employees were required to work outside of the scheduled nine- *102 to-five work day, Kronos automatically recorded such hours. In those weeks in which Kronos recorded Vitali as having clocked more than forty-five hours, she was paid overtime. For example, during the week of February 28, 2011, Kronos recorded that Vitali clocked 49.75 hours, and she was paid for four and three-quarters hours of overtime. According to Vitali, her work responsibilities also required her to work during her lunch breaks on average three to four times per week. The employees in her unit did not have specifically scheduled lunch breaks; instead, people took them “when they could.” Vitali “always” brought her lunch and “typically” ate it at her desk in her cubicle. While she was taking such breaks, people would bring her assignments that required prompt attention. Vitali provided numerous examples of specific individuals who would bring such assignments and the kinds of tasks that required her to do work during lunch. For example, she identified Carrie Noyes as someone who “would come to [her] with bank reconciliation items that she needed resolved right away for [the company’s comptroller and another high ranking manager].” Vitali also stated that she regularly observed others working during their lunch breaks, and she specifically identified such individuals. It is uncontested that Vitali never successfully used the Kronos drop down menu protocol to record the lunch time work she claims to have performed, 4 and that she did not receive credit for any such work toward the accrual of overtime. Had she been credited for the work, she would have received some additional overtime compensation (in those weeks in which her total worked hours exceeded forty). 5 5. The judge’s ruling. The judge concluded that with respect to Vitali’s uncorroborated claims to having worked regularly during lunch, “her deposition testimony to this effect is sufficient to create a jury question on [this issue].” However, he went on to rule in the company’s favor on other grounds. Specifically, he *103 concluded that Vitali had failed to produce evidence upon which reasonable jurors could conclude that the company knew or should have known that Vitali had engaged in uncredited overtime. In this regard, the judge deemed it critical that Vitali had failed to report her lunch time work in accordance with available procedures, even in the face of the company’s general policy against paying overtime except where employees had obtained prior approval. The judge also found it significant that — in contrast to some of the cases that Vitali had cited — there was no evidence here that the company had pressured Vitali not to report the hours for which she was seeking credit. Discussion. The payment of overtime is governed by G. L. c. 151, § 1A. That statute “was ‘intended to be essentially identical’ to the Fair Labor Standards Act of 1938 (FLSA), 29 U.S.C. § 207(a)(1) (2000).” Mullally v. Waste Mgmt. of Mass., Inc., 452 Mass. 526, 531 (2008), quoting from Swift v. AutoZone, Inc., 441 Mass. 443, 447 (2004). Accordingly, in interpreting the State law, we look to how the FLSA has been construed. See ibid. The case law has interpreted the FLSA in a manner that is highly protective of employee rights. As the United States Court of Appeals for the Second Circuit recently observed, “[i]n service of the [FLSA’s] remedial and humanitarian goals, the [United States] Supreme Court consistently has interpreted the [FLSA] liberally and afforded its protections exceptionally broad coverage.” Chao v. Gotham Registry, Inc., 514 F.3d 280, 285 (2d Cir. 2008). Pursuant to the FLSA, an employee must prove both that he incurred unpaid overtime work, and that the employer “had actual or constructive knowledge that he was working overtime.” Prime Communications, Inc. v. Sylvester, 34 Mass. App. 708, 709 (1993). 6 The knowledge inquiry requires an assessment of what the employer knew or should have known, and is to be made in view of the employer’s “duty ... to inquire into the conditions prevailing in his business.” Gulf King Shrimp Co. v. Wirtz, 407 F.2d 508, 512 (5th Cir. 1969) (quotation omitted). In other words: “In reviewing the extent of an employer’s awareness, a court *104 ‘need only inquire whether the circumstances . . . were such that the employer either had knowledge [of overtime hours being worked] or else had the opportunity through reasonable diligence to acquire knowledge.’ ” Reich v. Department of Conservation & Natural Resources, 28 F.3d 1076, 1082 (11th Cir. 1994), quoting from Gulf King Shrimp Co. v. Wirtz, supra. To the extent that an employee has reported his hours in accordance with the employer’s mandated timekeeping procedures, the employer’s knowledge of those hours is not in doubt. Thus, the cases concerning an employer’s knowledge all involve employee claims for unreported hours. In such cases, any failure by the employee to use prescribed timekeeping procedures is obviously a point in the employer’s favor. However, that failure is not fatal to the employee’s claim if he or she is able to marshal other proof that the employer had actual or constructive knowledge of the unpaid overtime. See, e.g., Holzapfel v. Newburgh, 145 F.3d 516, 524 (2d Cir. 1998) (“[0]nce an employer knows or has reason to know that an employee is working overtime, it cannot deny compensation even where the employee fails to claim overtime hours”). Thus, even where the employer has expressly prohibited overtime work, if it had reason to believe that such work was being done, “the employer cannot sit back and accept the benefits without compensating for them.” Reich, supra at 1082, quoting from 29 C.F.R. § 785.13. 7 Conversely, if the employee is unable to marshal proof that the employer knew or should have known of the overtime work, the employee cannot prevail. See Prime Communications, Inc., supra at 711, quoting from Forrester v. Roth’s I.G.A. Foodliner, Inc., 646 F.2d 413, 414 (9th Cir. 1981) (no FLSA liability “where an employer has no knowledge that an employee is engaging in overtime work and that employee fails to notify the employer or deliberately prevents the employer from acquiring knowledge of the overtime work”). We are mindful that, in reviewing the summary judgment record, we must consider not only any direct evidence of the em *105 ployer’s knowledge (actual or constructive), but also “all reasonable inferences” to be drawn from the evidence. Godfrey v. Globe Newspaper Co., 457 Mass. at 119. Indeed, an employer’s knowledge, like other “state of mind” inquiries, “is elusive and rarely is established by other than circumstantial evidence.” Blare v. Husky Injection Molding Sys. Boston, 419 Mass. 437, 439 (1995). Questions such as knowledge and intent often “require[ ] the jury to weigh the credibility of conflicting explanations.” Id. at 440. Thus, the determination of what a person knows or should have known under a specific factual situation is typically ill-suited for resolution by summary judgment. Riley v. Presnell, 409 Mass. 239, 247-248 (1991). Turning to the application of these principles here, we first examine whether the record creates a factual dispute regarding whether the company knew or should have known that Vitali did not take her full lunch breaks. We then turn to whether there was sufficient evidence in the summary judgment record that the company knew or should have known that Vitali was not receiving credit for such time. There was ample evidence upon which jurors could conclude that the company generally was aware that many of its employees sometimes worked during their allotted lunch breaks. For example, as discussed further below, when Kronos was first rolled out, the company’s payroll department received multiple employee inquiries about how to record lunch time work. 8 One of the company’s own affiants even stated that during the relevant period, she “typically worked through lunch” as a matter of mere personal preference. Indeed, the practice of employees working through lunch apparently became so pervasive that the company on several occasions had to remind employees that they were supposed to take at least a one-half hour lunch break. 9 Other than those periodic reminders, there is no evidence that the company sought to limit its employees from working during their lunch *106 breaks. This is hardly surprising given that work done during a lunch break costs the company no extra direct compensation (since the employees were already being paid for that time). 10 , In addition, there was evidence from which reasonable jurors could conclude that the company knew, or had reason to know, that Vitali in particular did not take her full one-hour lunch breaks. Unlike the typical overtime case where the extra work the employee claims to have performed was done off site, the alleged work here was done at her cubicle desk in an office setting. 11 It is uncontested that Vitali typically took lunch breaks at her desk, and the company concedes “that, at times, Ms. Vitali received various work-related assignments throughout her day” from her supervisor. 12 The company has not actually challenged Vitali’s specific averments as to the pressing nature of the assignments that she claims prevented her from taking her full lunches. 13 Especially given that the state of an employer’s knowledge is to *107 be assessed in light of its duty to inquire into the attendant working conditions, there was ample evidence on which jurors reasonably could have concluded that the company at least had reason to know that Vitali sometimes performed work during her lunch breaks. There was also evidence, discussed infra, that the company had actual knowledge of at least one occasion on which Vitali worked through her lunch break. 14 Seeking to avoid the implication that it had reason to know that Vitali was performing work during her lunch breaks, the company highlights that it had a sternly worded policy in place requiring all employees to obtain specific prior approval before working overtime. 15 See Newton v. Henderson, 47 F.3d 746, 749-750 (5th Cir. 1995) (employee cannot thwart clearly enforced policy against working overtime). This argument is unavailing for two reasons. First, the company has not shown that its policy requiring prior approval for overtime work had any application to employees performing work during their lunch breaks. Although lunch time work (like any other work done during the scheduled nine-to-five work day) counts toward the forty-hour threshold that needs to be crossed for overtime to be due, it does not itself constitute overtime. Thus, an hourly employee who worked through every lunch break would still not be entitled to any overtime unless she performed additional work outside of the normal nine-to-five work day. Notably, in reminding employees of the need to seek prior approval for overtime work, the written instructions that the company provided for using Kronos specifically equated “working overtime” with “coming in early or staying late.” 16 The company was free to adopt a policy requiring employees to obtain *108 prior approval before performing work on their lunch breaks; it simply did not do so. 17 Second, there was evidence in the summary judgment record that the company’s policy of requiring specific prior approval for overtime was honored in the breach. For example, one of the company’s own affiants stated that her supervisor had given her “general blanket approval for overtime” when her department was “unusually busy.” Where an employer in practice fails to enforce a formal employment policy limiting overtime work, a jury could infer that the employer knew or should have known that employees were engaged in unauthorized overtime notwithstanding the existence of such a policy. See Reich, 28 F.3d at 1083 (recognizing that employer must do more than “simply continue to apprise [the employees]” of policy against working overtime). In addition, the company argues that even if it had reason to know that Vitali at times was not taking her full allotted lunch breaks, it still had no reason to know that she was not reporting this lunch time work. In making this argument, the company asserts that employees were well informed of the way in which hours worked during lunch were to be recorded in Kronos. The company maintains that it was fair and appropriate to assume that Vitali would have reported any lunch time work through the means that the company made available, especially where employees were required to attest to the accuracy of their recorded time. In short, the company contends that because Vitali failed to comply with reasonable reporting procedures, her case fails as a matter of law. See White v. Baptist Memorial Health Care Corp., 699 F.3d 869, 876 (6th Cir. 2012) (“Under the FLSA, if an employer establishes a reasonable process for an employee to report uncompensated work time the employer is not liable for nonpayment if the employee fails to follow the established process”). 18 *109 There are several problems with this argument. To begin with, “[t]he FLSA makes clear that employers, not employees, bear the ultimate responsibility for ensuring that employee time sheets are an accurate record of all hours worked by employees.” Skelton v. American Intercontinental Univ. Online, 382 F. Supp. 2d 1068, 1071 (N.D. Ill. 2005). Moreover, “an employer’s duty under the FLSA to maintain accurate records of its employees’ hours is non-delegable.” Kuebel v. Black & Decker Inc., 643 F.3d 352, 363 (2d Cir. 2011). The company has not shown, as a matter of law, that it has satisfied its timekeeping responsibilities here. Although the company maintains that it instructed employees to record lunch time work and provided them an accessible and transparent means of doing so, there plainly was evidence on which reasonable jurors could have concluded otherwise. Explaining this requires a close examination of the instructions that employees received when Kronos was implemented. During the rollout of Kronos, employees were given two sets of written instructions: a five-page manual and an instructional electronic mail message (email) announcing the “Good News” that Kronos was being implemented. These documents provided discordant advice on the lunch break issue. The manual instructed employees to clock out when they began their lunch breaks and to clock back in when they returned, despite the fact that, as noted, the version of Kronos that the company initially had installed did not have that functionality. 19 Had Kronos included that feature (as it later did following Vitali’s departure, see note 8, supra), then the time each hourly employee clocked would have been the same as the time they actually worked (and hence there would have been no need for employees separately to record lunch time work or for the company to make assumptions regarding whether such work took place). The “Good News” email, meanwhile, instructed that hourly employees “do not have to punch in and out for lunch.” What is particularly telling, however, is what the email did not say. Nowhere does it state that employees were required to account for any time they worked during their lunch breaks. Nor did the email explain that employees’ failure to do so might affect their *110 overtime. In fact, the email did not even explain how employees could record their lunch time work in the Kronos system except for a cryptic notation that “[i]f you are to work through a lunch, please use the ‘hours worked’ code and add the amount of in time worked in increments of .25 only.” In distributing the instruction manual and email, the company told employees that if they had problems or follow-up questions, they should contact the payroll department for assistance. Although the email raised the possibility of group training sessions being held, there is no evidence in the record that any were provided. Individual employees in fact did follow up with the payroll department to complain that Kronos was “not user friendly.” Multiple employees specifically asked about what they should do about recording lunch time work. The company’s payroll supervisor stated that she generally responded to these inquiries by trying to instruct the inquiring individual on the appropriate procedure for recording such time, the drop down menu protocol referenced supra. 20 The payroll supervisor conceded that she did not necessarily provide such instruction to all individuals who inquired about this subject, and she was unable to explain why she did this in some cases but not others. 21 In an email exchange, Vitali herself puzzled over what to do about recording time worked during her lunch hour. Specifically, on March 15, 2010, a month after Kronos was implemented, Vitali sent an email inquiry to the payroll department (as instructed). Vitali noted that she was unable to take lunch that day, and she sought advice on how to “mark” that in Kronos. Her email was forwarded to payroll supervisor Juppe, who responded that “[i]f your physically worked hours for this week [are] over 40 hours we can discuss how the lunch time for today should be recorded.” In response, Vitali stated that it was just that one day *111 that week that she had to work through lunch, and she requested that Juppe tell her what she needed to do. Juppe responded that “[t]here is nothing you need to do for this as your total hours for the week do not exceed 45.” 22 This was the last interchange that Vitali and Juppe had on the subject: there is no evidence that Juppe provided Vitali any guidance about how to report time worked during lunch in weeks where it could make a difference, nor that Vitali made any further inquiries about this. In sum, reasonable jurors could make the following factual findings based on the summary judgment record. When the company moved to Kronos, it did not require that employees record any lunch time work or even explain to them how their not recording that time might affect their over-all compensation. Although the company did provide a method through which employees could record lunch time work, the written instructions that it provided about doing so were contradictory, confusing, and incomplete. Moreover, at least by the time its payroll department received the follow-up inquiries from individual employees regarding the use of Kronos, the company at least had reason to know both that many employees were performing work during their lunch breaks and that they were confused as to what to do about recording such time. The company did not provide training to Vitali on how to record lunch time work even when she specifically had sought advice on the issue. Instead, the payroll department advised her that there was nothing that she needed to do at that time, while leaving her in the dark as to what the proper procedure would be where her lunch time work could make a difference. 23 In short, armed with at least constructive knowledge that employees were undertaking lunch time work that should have been credited toward overtime, the company went ahead and assumed in its favor that employees were not performing any such work except where they separately reported it through a process that Vitali was never trained in, or even told to use. Under these circumstances, the judge erred in concluding that the company was entitled to *112 judgment as a matter of law. 24 Judgment reversed. Footnotes [1] The complaint also alleged contract and quantum meruit claims, but Vitali — who was an at-will employee — abandoned such claims before the summary judgment motion was resolved. [2] Prior to that date, the company used a paper-based system. Vitali initially asserted that she was shortchanged overtime under that system as well, but she has since abandoned such claims because of the applicable statute of limitations. [3] The time code for “worked hours” (also referenced in the record as “hours worked”) was distinct from the one for “regular work hours.” [4] According to her deposition testimony, Vitali did try to use this protocol on one occasion and was unable to do so. It is not clear exactly when this is alleged to have occurred. [5] Thus, for example, if Vitali were credited for doing a total of two hours of lunch time work in the week that she clocked 49.75 hours in Kronos, she would have been entitled to six and three-quarters hours of overtime (since her clocked time would have included only three hours of lunch breaks, not the five that the company assumed). [6] The cases have consistently so held. Nevertheless, Vitali argues that under G. L. c. 151, § 1A, an employer should be liable regardless of whether it knew or should have known of the overtime. In the alternative, Vitali argues that the employer should have to bear the burden of proof regarding its lack of knowledge. Putting aside that these arguments were not raised in the trial court (and therefore have been waived), Vitali has provided no reason why G. L. c. 151, § 1A, should be interpreted at odds with the “essentially identical” ELSA. [7] As the judge recognized, some cases that have found sufficient evidence of employer knowledge of unpaid overtime (in the face of timekeeping records to the contrary) rest in part on evidence that the employer pressured employees to underreport their time. See, e.g., Kuebel v. Black & Decker Inc., 643 F.3d 352, 363-364 (2d Cir. 2011) (employees told not to record their overtime). However, none of these cases holds or even suggests that such bad faith conduct is a prerequisite. [8] Eventually, Kronos was upgraded so that employees could —• and were required to — log out for lunch and log back in when they returned (thus allowing more accurate reporting of the length of the lunch breaks that employees took). The records generated after this upgrade showed a widespread practice of employees not taking their full lunch breaks. Although the Kronos system upgrade took place after Vitali had left the company’s employ, there was deposition testimony that employee lunch patterns were the same both before and after the upgrade. [9] Pursuant to G. L. c. 149, § 100, employers are required to provide employees who work at least six-hour shifts a one-half hour lunch break. [10] In pointing out that the company did not have an incentive to discourage lunch time work, we do not mean to suggest that the company therefore was acting in bad faith. [11] In the off-site context, a supervisor may well have little basis for knowing how many hours an employee has worked except to the extent the employee reports them. This tends to put a premium — for both employer and employee — on employee compliance with whatever reporting systems are in place. Nonetheless, an employee’s failure to report off-site overtime, even while attesting to the accuracy of his time records, is not fatal if the employee can produce evidence that the employer knew or should have known of the overtime. Reich, 28 F.3d at 1084. See 29 C.F.R. § 785.12 (making explicit that employer that knows or has reason to know of work performed away from job site must compensate for that time). [12] We note that the Federal regulations issued pursuant to the FLSA address when nominal meal breaks are sufficiently interrupted by work demands to be considered compensable work time. See 29 C.F.R. § 785.19; Beasley v. Hillcrest Med. Center, 78 Fed. Appx. 67, 69 (10th Cir. 2003) (analysis under 29 C.F.R. § 785.19 focuses on “whether the degree of interruption caused [the employees] to spend their meal periods primarily for [the employer’s] benefit”). Neither party has addressed the potential relevance of the FLSA regulations to this case, and we do not rely on them. [13] The company accurately points out that Vitali did not produce evidence to “demonstrate[ ] that a supervisor or other individual ever specifically asked Ms. Vitali to perform such work during her lunch break.” However, such proof is unnecessary, since “an employer can be charged with constructive knowledge even when an employee has not alleged a supervisor’s direct knowledge.” Allen v. Board of Pub. Educ., 495 F.3d 1306, 1321 (11th Cir. 2007). If the company knew or had reason to know that Vitali was performing the assigned tasks during her lunch breaks, whether anyone specifically directed her to do the work at that time is beside the point. [14] A month into the rollout of Kronos, Vitali informed the payroll department that she was unable to take lunch on a particular day (thus providing the company with direct knowledge that she had performed lunch time work that was not recorded in Kronos). Although Vitali stated that this was the only day that week in which she worked through lunch, she never represented to the company that it was the only time this had ever or would ever occur. In addition, while the company suggests that Vitali admitted that she took her full one-hour lunch breaks by commenting (in response to criticism about her job performance) that “I am working from the time I come in, excluding lunch until the time I leave,” reasonable jurors would not be required to attach such import to that comment. [15] That policy edict was plainly stated in the employment manual that all employees received, and it was widely disseminated to employees through other means as well. [16] The company made no showing that Vitali failed to report her work done before nine or after five. To the contrary, it attempts to make use of the fact that she knew how to report the overtime for which she was paid. [17] In fact, the company’s information technology (IT) unit instituted a strict policy that required members of its support services team to obtain a supervisor’s approval for any work done during a lunch break. This was set forth in a memorandum that also explained the relationship between working through lunch and the accrual of overtime hours. As the judge recognized, “[tjhere is no evidence in the summary judgment record to suggest that Vitali, who did not work in the IT department, actually received or was aware of [this] memorandum.” [18] The company acknowledges that there would be a different result if Vitali were able to show that it pressured employees into underreporting their hours *109 (see note 7, supra), but it accurately points out that there was no such evidence presented here. [19] Confusingly, the instruction manual included a note that stated in pertinent part: “In most regions, lunch breaks are auto deducted whether you clock out or not • — • check with your manager for instruction.” [20] The company produced affidavits from two current employees who stated that they were told how to navigate the protocol for recording lunch time work and who averred that they were “not aware of any overtime for which [they had] not been properly paid.” Of course, the fact that the company successfully may have trained some employees about how to record lunch time work says nothing about the training it provided to Vitali. [21] Juppe’s deposition answers suggest that she may have drawn a distinction between time that employees spent working “through” a lunch (as to which she would tell employees they should use the drop down menu protocol) and time that employees spent addressing “quick work assignment[s]” that they were asked to do while taking lunch at their desks (as to which she responded that she was “not sure” what employees should do to record their time). [22] Although the judge concluded that Juppe’s response “would not have been misleading,” he recognized that in fact it “was mathematically inaccurate.” [23] In recounting the summary judgment record in the light most favorable to Vitali for purposes of the current appeal, we of course do not mean to suggest that the company necessarily knew or should have known that Vitali was due unpaid overtime. That question will be for the jury to decide based on the trial record. [24] Because the judge allowed the company’s motion for summary judgment, he had no occasion to consider whether the case properly could be maintained as a class action. We express no view on that issue.