Caradigm USA LLC v. PruittHealth, Inc.

ORDER HONORABLE STEVE C. JONES, UNITED STATES DISTRICT JUDGE Data analytics and computer software create efficiencies in many industries, nursing home providers included. That’s why defendant Pruitt Health, Inc., one such provider in the southeast, contracted with Caradigm USA LLC to condense several different electronic medical- records systems into one longitudinal patient summary. After a few months, however, Pruitt walked away from its relationship with Caradigm over what it perceived as Cara-digm’s inability to perform essential services. That led to this breach of contract suit against Pruitt and ultimately to the dueling summary judgment motions (docs. 65 & 67)1 and motion to exclude expert testimony now before this Court. Doc. 75. I. BACKGROUND Pruitt “is a privately held, for-profit healthcare provider that among other things operates nursing homes.” Doc. 73 at 2 (Caradigm’s Statement of Undisputed Fact). “Caradigm provides software services to healthcare companies.” Id. at 1. When Pruitt began looking for an IT partner to help it synthesize patient data stored in multiple software systems, the “Caradigm Intelligence Platform” (CIP) piqued its interest. Doc. 67-1 at 1 (Pruitt’s Motion for Partial Summary Judgment Opening Brief). CIP, according to Cara-digm, — 1178 — aggregates and normalizes clinical and financial data from across the care continuum, and allows for data re-use for a more adaptable and strategic approach. Unlike a typical data warehouse, CIP aggregates data and makes it available in near real time, allowing insights within a workflow. The key point is that having all the data in one place creates an entirely new ‘asset’ that functions as a platform to ‘dive off from’ and or build broader tools upon. Doc. 1-3 at 6 (Statement of Work) (SOW). Those features, and CIP’s attendant ability to “[p]rovide ... Pruitt with a comprehensive patient summary view based on patient information” from disparate systems, led Pruitt to contract with Caradigm “for the customization, implementation, and licensing of CIP.” Doc. 73 at 2. A. The Parties’ Contract Three documents, all executed on June 25, 2013, comprise the parties’ contract (collectively, the “Agreement”): a “Cloud Services Agreement” (CSA) (doc. 1-1); an “Order for a set of services” (Order) (doc. 1-2); and the SOW (doc. 1-3). Doc. 73 at 2. The CSA sets forth the legal framework that governs the parties’ relationship, the Order2 documents the precise services and goods Pruitt purchased pursuant to the CSA and their costs, and the SOW details the nuts and bolts of how Pruitt and Cara-digm would integrate a complicated software system like CIP into Pruitt’s business. All three documents contemplate a rough divide of the relationship into two halves—before Pruitt used CIP “to process actual patient data in a live production environment,” and after. Doc. 1-1 at 7. The CSA brands that fulcrum as the “First Productive Use.” Id. Implicitly recognizing the significance of that moment, the Order invoiced one set of fees to be paid monthly prior to “go live,” and another, much more expensive set, for after. See doc. 1-2 at 2. The SOW similarly acknowledged a preparation-intensive “testing” phase that would precede “First Productive Use.”3 Doc. 1-3 at 23. Substantively, the CSA gave Pruitt “the right to access and use ... Cloud Services,” i.e., CIP. Doc. 1-1 at 2. Because Pruitt purchased a development license (the right to itself customize CIP)—for which it paid $20,000 every month beginning the day it executed the CSA and Order (doc. 1-2 at 2)—the CSA allowed Pruitt to not only access CIP, but also to “build Customer Use Cases4 and configure Data Feeds5 for those” cases. Doc. 1-1 at — 1179 — 2 (footnotes added). “Following [CIP’s] Installation” (i.e., configuring Cloud Services for Pruitt to test or use, id. at 7), Pruitt could “test the Cloud Services to determine whether they perform[ed] in material conformance with the Documentation” (technical guides and user manuals that Caradigm provided). Id. at 3. That particular “testing period” expired at “the earlier of: (a) First Productive Use of those Cloud Services; (b) the date on which those Cloud Services materially conform to the Documentation; or (c) 35 days from Installation of those Cloud Services without Car-adigm having received written notice from Customer of any material nonconfor-mance.” Id. For testing Customer Use Cases (a subgroup of Cloud Services), something only Pruitt as the end user and sometimes designer of those use cases could do, the CSA obligated Pruitt to “notify Caradigm when testing [was] satisfactorily completed and ... Caradigm [could] move the Customer Use Case to the production environment.” Id. The Order-—-“made as of June 25th, 2013 (“Order Effective Date”) pursuant to the [CSA] ”—meanwhile, reveals that Pruitt, in addition to the $20,000/month CIP development license, purchased two other “Cloud Services:” (1) a CIP subscription for $64,649 per month, with payments “starting on First Productive Use,” and (2) a “Pre-First Productive Use Hosting Fee” of $4,000, charged monthly from the CIP “Platform Implementation kick-off6 until First Productive Use.” Doc. 1-2 at 2 (emphasis in original, footnote added). Pruitt also ordered several professional services from Caradigm related to installing CIP on Pruitt computers and setting up Data Feed and Use Cases installations. Id. Together, Caradigm invoiced Pruitt $650,000 after discounts for those services. Id. Back to the CSA’s legal framework. Its “Term & Termination” provision defines the duration of the parties’ agreement and the circumstances under which unilateral termination may occur. The “Term” portion bifurcates the length of agreement into an “Initial Term” and possible “Renewal Terms.” Doc. 1-1 at 3. The parties’ “Agreement,” says the CSA, “will run for an Initial Term ending five years after the date of [CIP’s] First Productive Use.” Id. (emphasis added). Renewal Terms are twelve month extensions of the Initial Term that occur if neither “party provides written notice of non-renewal at least 90 days before the end of the then-current Term.” Id The “Termination” provision states that “[e]ither party may terminate [the CSA] or any Order if the other party commits a material breach of [the CSA] or the applicable Order that remains uncured for 90 days after written notice of such breach.” Id. No other provision in the CSA, Order, or SOW addresses the Agreement’s duration or termination. The CSA also addressed CIP warranties.7 Although Caradigm warranted that its “Cloud Services [would] perform substantially as described in their Documentation,” it disclaimed all “other express or implied warranties ... with respect to software or services.” Doc. 1-1 at 4-5. In particular, it disclaimed any warranty “with respect to ... [the] correctness or reliability of patient information (including patient record matching)” in addition to merchantability and fitness for a particular — 1180 — purpose warranties.8 Id. at 5. Finally, the CSA made clear that it and any referenced documentation “constitute[d] the entire agreement of the parties ... [and] spper-sede[d] all other prior or concurrent agreements.” Id. at 6. B. Contractual Performance After some hiccups getting started (related primarily to turnover in key positions at Pruitt, see doc. 73 at 4), the parties began in July 2014 to implement CIP. Part of that process involved Pruitt sending batches of patient data to Caradigm so the latter could test its systems’ ability to “match” patient files from various sources. ■That “patient matching”—a “complex task ... whereby records associated with a patient in source system A (of which there can be many per patient) are identified and linked to all the records associated with the same patient in source systems B, C, D, E, and F,” doc. 67-1 at 49—under-girded (in Pruitt’s eyes anyway) CIP’s usefulness to Pruitt. Doc. 72 at 3-10. The December 2014 results of that initial testing showed a match success rate that, regardless of the precise percentage of records matched, did not satisfy Pruitt. Doc. 72 at 27-30. The parties dispute what happened thereafter and the relevance of what indisputably did occur (see, e.g., id. at 31), but neither side contests that additional discussions about how best to proceed with CIP'implementation took place. Then, on January 15, 2015, Pruitt issued an “official status report” for the project that reflected Caradigm’s belief “that there is a sufficient level of matching ... in [its] systems to produce a viable matching result,” and that “[s]ince Caradigm is not a true EMPI system, like NextGate (who Caradigm recommends), patient matching is limited to 100% matching of specific field requirements.” Doc. 73 at 17. Four days later Pruitt’s executive leadership met and decided to pull the plug on CIP implementation. Doc. 73 at 27-28. Pri- or to that meeting, Pruitt never notified Caradigm “of dissatisfaction with Cara-digm’s performance.” Id. at 28. Finally, on January 29, 2015, Pruitt’s IT head, Dan Martin, told Caradigm employee Tina Mirkheshti in a phone call that'Pruitt “will not be able to move forward with CIP.” Id. at 30. Neither party disputes that Martin told Mirkheshti that part of the reason for discontinuing CIP was “that [Pruitt was] going to do another [electronic medical record system].” Id. at 31. Mirkheshti expressed her surprise in an email to Pruitt’s CIP project manager on February 10, 2015. Doc. 73 at 31. A week later, Pruitt’s Marty Meighan, Senior YP of Supply Chain Management, emailed Mirkheshti — 1181 — to make sure we all [are] of the same understanding regarding the future of this project. As far as I have been informed, we have suspended all efforts dedicated to this project and that includes utilization of your services. Therefore the contract you have with PruittHealth is no longer valid. In addition regarding payments, we will make our final payment for January date(s) of service, and this will fulfill our economic commitment under the agreement. Doe. 55-26 at 2. Caradigm’s CEO responded a month later: Under the terms of the Agreement, UHS-Pruitt does not have a unilateral right to terminate the Agreement during the Agreement’s five-year initial term (see Section 12 of the Agreement). Your purported termination would therefore constitute a breach of our Agreement, and this breach would result in the acceleration of payments owed to Cara-digm for the contracted five-year period. Caradigm stands ready, willing, and able to continue to fulfill our obligations under the Agreement. Accordingly, we would urge UHS-Pruitt to proceed under our contract as previously agreed. While we regret your apparent intention to breach our Agreement, we ■ will of course respect your instructions to hold off on implementation work on your behalf at this time. Doc. 1-6 at 2 (letter sent on' March 16, 2015). Pruitt never responded to that letter. The kick off meeting was never scheduled. And neither party ever took another step towards CIP implementation at Pruitt. Caradigm filed this action on July 14, 2015 asserting that Pruitt breached the parties’ contract by refusing “to pay any further fees to Caradigm.” Doc. 1 at 8. Pruitt’s “unilateral suspension of [CIP] implementation and its refusal to allow Cara-digm to perform under the Agreement,” says Caradigm, failed to comply with the CSA’s termination provision, and “constitute[d] an anticipatory breach, entitling Caradigm to the full and accelerated .value of the Agreement.” Id. Less than a month after the complaint’s filing, and before Pruitt answered or appeared, its attorneys sent Caradigm a letter outlining the arguments that would later undergird its defenses and summary judgment position. Doc. 60-33 at 1-2. The letter also purported to serve as “written notice of material breach,” though it did not offer Caradigm the opportunity to cure. Id. at 3. Cara-digm, having already filed suit, had no interest in conversation at that time. A month after Pruitt’s post-suit filing notice letter, it timely answered and counterclaimed. Doc. 7. It argues that Cardigm in fact breached (1) the agreement by failing to deliver on its patient matching promises, (2) its“workmanlike services” warranty, and (3) the implied duty of good faith and fair dealing. Id. at 30. After uneventful discovery, both parties filed motions for summary judgment. Docs. 65 & 67. Caradigm argues that Pruitt’s unilateral termination of the Agreement failed to comply with its termination clause by never providing written notice of an alleged material breach and an opportunity to cure. Doc. 65-1 at -4. “[E]ven apart from” that notice failure, says Caradigm, Pruitt breached the parties’ contract by anticipatorily repudiating it before the time for Caradigm’s performance ran out. Id. at 4-5. Pruitt’s motion, on the other hand, spends no time on its own breach claim. Instead, it asks the Court for a quasi-declaratory judgment (something its counterclaim never mentions) holding (a) that the [CSA] should be construed as either (i) permitting Pruitt to end implementation prior to com1182 meneement of the “Initial Term” upon full payment of monthly subscription fees up through the date of project cancellation; or (ii) establishing a condition subsequent of satisfactory testing and First Productive Use, such that the inquiry for the jury is whether Pruitt “fairly exercised” its discretion under the contract; or, in the alternative, (b) that Caradigm’s recoverable damages, if any, are limited to development license fees and professional services fees that would have been incurred through November 8, 2015, minus saved costs. Doc. 67-1 at 25. II. STANDARD OF REVIEW Summary judgment is appropriate when the moving party establishes that, based upon the evidence presented, “there is no genuine dispute as to hny material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). “[T]he requirement that a dispute be ‘genuine’ means simply that there must be more than some metaphysical doubt as to the material facts.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 261, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986) (citations and internal quotation marks omitted). The court views the record and draws all factual inferences in the light most favorable to the non-movantf, in this case, MVG]. Carlson v. FedEx Ground Package Sys., Inc., 787 F.3d 1313, 1317 (11th Cir. 2015). “If reasonable minds could differ on the inferences arising from undisputed facts, then a court should deny summary judgment.” Id. at 1318 (quoting Allen v. Tyson Foods, Inc., 121 F.3d 642, 646 (11th Cir. 1997)). Dean-Mitchell v. Reese, 837 F.3d 1107, 1111-12 (11th Cir. 2016). III. ANALYSIS Before turning to the two summary judgment motions, the Court must first address Caradigm’s motion to exclude the testimony of Pruitt’s expert, Cheri Kane. Doc. 75. A. Motion to Exclude Pruitt’s summary' judgment arguments, as noted above, center on its belief that the CSA allowed it to unilaterally terminate prior to First Productive Use if Cara-digm’s patient matching performance dissatisfied Pruitt. To bolster that position, Pruitt hired Kane, a Managing Director at PricewaterhouseCoopers with a background in healthcare administration, to provide an opinion on patient matching’s role in CIP’s performance. Doc. 75-1. In particular, Kane opined that: —The agreement(s) entered into between the parties make specific references to capabilities that require advanced patient matching. —Without sophisticated patient matching, the CIP solution could not create a unified patient intelligence and reporting platform to support the Use Cases laid out in the SOW. —Because the parties never completed a successful testing phase, First Productive Use was never achieved. Id. at 15. Caradigm contends that Kane’s testimony must be excluded because (1) she is “not a patient matching expert and did not analyze either Caradigm’s software or Pruitt’s data in any way;” (2) her “opinions ultimately amount to ... lay interpretations of the Agreement’s terms;” and (3) her “opinions rely ... upon assumptions she made at the instruction of Pruitt’s counsel.” Doc. 75 at 8-9. Pruitt disputes all three exclusion grounds. Doc. 81. The Court need not decide whether to exclude Kane’s testimony because even if admitted, it (1) creates no genuine disputes of fact that preclude summary judgment, — 1183 — and (2) has no relevance to the current motions’ core legal questions. Without completely rehashing what is discussed below in greater detail, the Court finds that the Agreement contains an unambiguous termination clause that applied at the time Pruitt walked away and controlled how unilateral termination could occur. Whether or not Caradigm’s patient matching performance constituted a breach of the Agreement, Pruitt never complied with that clause. More importantly for purposes of Cara-digm’s exclusion motion, none of Kane’s opinions has any bearing on either of those conclusions. As she notes, the Agreement in fact makes specific reference to patient matching, and likely also to “capabilities that require” that feature. Doc. 75-1 at 15. And assume that CIP could not perform without advanced patient matching, and that the parties never reached First Productive Use (a point both parties concede). None of those facts bears on the question driving the Court’s breach analysis: whether the termination provision applies to all unilateral attempts to end the Agreement, or whether it only applies after the parties achieve First Productive Use. Take the Agreement’s explicit reference to patient matching. Does that affect the termination clause’s applicability? No. If anything, it supports Pruitt’s argument and counterclaim that Caradigm breached the Agreement by providing pedestrian patient matching. But defining the contours of what constitutes a material breach says nothing about the circumstances under which a party may terminate. Likewise for CIP needing advanced patient matching in order to provide the efficiencies Caradigm promised. How does that bear on the circumstances under which a party could terminate the Agreement? It doesn’t. Like contractual mentions of patient matching, CIP’s patient matching needs at best suggests that Car-adigm breached before Pruitt unilaterally terminated. Also like those mentions, however, Kane’s second opinion does not inform the termination clause question. The notion that the parties never reached First Productive Use because of testing failures trods the same path. Taking that fact as gospel moves the Court no closer to deciding when the Agreement began, or under what circumstances the parties could unilaterally terminate. Because the same can be said for all of Kane’s opinions, her testimony is irrelevant for current purposes. Caradigm’s motion to exclude therefore is DENIED AS MOOT.10 Doc. 65. B. Summary Judgment Although the parties filed dueling summary judgment motions, Pruitt’s does not address its counterclaims. Instead, it argues for Pruitt’s preferred construction of the Agreement. See doc. 67-1 at 3. The following discussion accordingly addresses the parties’ competing visions. Those in turn dictate the viability of their respective claims.11 Nevertheless, for ease of analysis, the Court frames that contract construction dispute using Caradigm’s breach claim as an analytical catalyst. 1. Caradigm’s Breach Claim Caradigm’s claim’s success hinges on whether Pruitt breached the Agreement by walking away from CIP implementation. That question presents three preliminary issues: (1) at what point did the Agreement’s termination clause kick in— — 1184 — at the its execution, or when the parties reached First Productive Use? (2) Once applicable, did that clause provide the only unilateral off-ramp? And (3) if it applied as of execution, did Pruitt comply with its written notice and opportunity to cure requirements (and, possibly, did Caradigm commit a material breach)? If First Productive Use activated an exclusive termination clause, under what circumstances, if any, could a party walk away before that event occurred? Caradigm believes that the CSA’s “Term and Termination” provision applied from the moment the contract was executed, and contains the only avenue by which a party could unilaterally terminate. Pruitt, on the other hand, believes that provision only applied after First Productive Use. Prior to that point, Pruitt contends that it could terminate if CIP’s performance left it subjectively dissatisfied. “By the Agreement’s plain terms,” says Caradigm, the parties could terminate their contract in “one of only two ways.” Doc. 65-1 at 17. First, they could mutually agree to walk away. Doc. 1-1 at 4. No one contends that that occurred. So, urges Caradigm, Pruitt had one path to termination: give Caradigm written notice of its material breach and ninety days to cure. If no timely cure occurred, Pruitt could unilaterally end the relationship. Doc. 65-1 at 17. Whether or not Caradigm materially breached, Pruitt, insists Caradigm, never “provided the required notice.” Doc. 65-1 at 18. Meighan’s February 17, 2015 email cannot qualify because it “makes absolutely no assertion that Caradigm has done anything wrong with respect to the Agreement.” Id And the August 2015 letter from Pruitt’s attorney comes up short despite detailing Caradigm’s alleged breach because it came after this action’s commencement. Id. To Caradigm, Pruitt’s walkaway therefore itself constituted a material breach that obviated Caradigm’s own performance and justifies summary judgment in its favor. Id. What’s more, notes Caradigm, any patient matching performance issue cannot justify Pruitt’s unilateral termination because the time for Caradigm’s performance had yet to pass. Doc. 65-1 at 20-21. Indeed, the parties never even held the kick off meeting, much less began anything approaching CIP deployment, before Pruitt walked away. Id at 22. That amounts to anticipatory repudiation, and that “absolved [Caradigm] from remedying the situations of which [Pruitt] complaints].” Id at 21 (first alteration added, remainder in original) (quoting Clark v. Cox, 179 Ga.App. 437, 438, 347 S.E.2d 4 (1986)). Pruitt sees the Agreement’s termination clause a bit differently. Most importantly, it believes, that the Agreement “did not expressly preclude termination other than in the event of an uncured material breach, and it plainly did not do so prior to First Productive Use.” Doc. 74 at 13 (quotes and alterations omitted). Caradigm’s construction to that effect neuters “Pruitt’s contractual right to ‘test’ [CIP], and to ensure its satisfaction before moving to First Productive Use.” Id at 14 (emphasis in original). By contrast, “[r]eading the notice- and-cure clause in context,” as Pruitt urges, “makes uncured material breach a prerequisite for contract termination only after satisfactory testing, First Productive Use; and, thus, commencement of the five-year ‘Initial Term.’ ” Id. at 14-15 (emphasis in original). Regardless, says Pruitt, even if the Court accepts Caradigm’s proposed construction, Pruitt “substantially complied with [the] termination provision’s notice requirement.” Doc. 74 at 15 (emphasis omitted). “Taken together,” the January 29, 2015 phone conversation between Mar — 1185 — tin and Mirkheshti and Meighan’s February 17, 2015 “left little doubt as to the source of Pruitt’s dissatisfaction” and thus complied with the Agreement’s notice requirement. Because Caradigm never cured within 90 days thereafter, Pruitt insists its termination complied with the contract. Id. at 16. If not, the August 10, 2015 letter from its attorney left zero doubt as to the source of Pruitt’s dissatisfaction and its intent to terminate. Doc. 74 at 16. It, too, therefore served as contractually compliant notice that submarines Caradigm’s bid for summary judgment. Id. Pruitt also rejects Caradigm’s anticipatory repudiation argument because, it says, Caradigm cannot recover “unless it was capable of performing.” Doc. 74 at 20. And that, at a minimum, is a “disputed issue[] of fact.” Id. Relatedly, Pruitt insists that its “dissatisfaction was not with how Caradigm might perform; it was how Caradigm actually performed when given an opportunity.” Id. at 21. Pruitt therefore could not anticipatorily repudiate—Cara-digm had in effect already breached. Id. The Court first considers the parties’ interpretive glosses on the Agreement, then turns to Caradigm’s anticipatory repudiation argument. i. Termination Clause—Activation, Timing, and Scope To decide when and under what circumstances the termination clause applied, the Court must interpret the Agreement. That analytical endeavor involves “a question of law, unless the contract language presents an ambiguity that cannot be resolved by the rules of construction.” Greenberg Farrow Architecture, Inc. v. JMLS 1422, LLC, 339 Ga.App. 325, 329, 791 S.E.2d 635 (2016).12 The cardinal rule of construction is to ascertain the intent of the parties. Where the contract terms are clear and unambiguous, the court will look to that alone to find the true intent of the parties. To determine the intent of the parties, all the contract terms must be considered together in arriving at the construction of any part, and a construction upholding the contract in whole and every part is preferred. When the language employed by the parties in their contract is plain, unambiguous, and capable of only one reasonable interpretation[,] the language used must be afforded its literal meaning and plain ordinary words .given their usual significance.... An ambiguity is defined as duplicity, indistinctness, an uncertainty of meaning or expression used in a written instrument, and also signifies of doubtful or uncertain nature; wanting clearness or definiteness; difficult to comprehend or distinguish; of doubtful purport; open to various interpretations. Where ambiguities exist, the court may look outside the written terms of the contract and consider all the surrounding circumstances to determine the parties’ intent. Parol evidence may not be considered unless the written instrument is ambiguous. Id. The Court begins with the Agreement’s “Term and Termination” provision. “Th[e] Agreement,” it says, “will run for an Initial Term [to] end five years after the date of First Productive Use,” and “will automatically renew for successive 12-month Renewal Terms unless either party provides written notice of non-renewal.” Doc. 1-1 at 3 (emphasis added). “Either party may — 1186 — terminate th[e] Agreement or any Order if the other party commits a material breach ... that remains uncured for 90 days after written notice of such breach.” Id. Neither the “Term and Termination” provision, nor any other portion of the Agreement, expressly contains another procedure by which a party can unilaterally terminate.13 ■ Pruitt insists that the Initial Term began'—and thus the termination clause awakened—at First Productive Use, which never occurred. That contention, however, implies a contradiction: that the Agreement required performance by both parties ($20,000+ per month in payments from Pruitt, installation of CIP and granting of a development license by Caradigm) before what it defined as the Initial Term’s start date. But how can a contract require performance, yet not be operational? If somehow the termination clause remains dormant despite the parties performing under the Agreement, what other provisions, if any, also remain deactivated prior to First Productive Use? Do warranty disclaimers apply? What about, for example, the Agreement’s choice of law provision? If all of those provisions apply prior to First Productive Use, why does the “Term and Termination” provision not? The text of the Agreement offers no answer to those questions. Pruitt makes much of the Intial Term’s inclusion of an end date, but not a clear beginning. Doc. 67-1 at 19 (“[T]he limitation on the right to terminate only upon written notice of material breach and a failure to cure within ninety days would ‘set in’ upon First Productive Use and remain in place for five years after that date.”). Despite the “Term and Termination” provision’s apparent silence on that point, the Agreement as a whole, see Greenberg Farrow Architecture, 339 Ga.App. at 329, 791 S.E.2d 635 (“all the contract terms must be considered together in arriving at the construction of any part”), provides unambiguous guidance. See In re Rail Freight Fuel Surcharge Antitrust Litig.-MDL No. 1869, 725 F.3d 244, 254 (D.C. Cir. 2013) (the “absence of evidence is not evidence of absence”). In that vein, the Order and the CSA cannot be read separately. As both Cara-digm and Pruitt acknowledge, they form two legs of the tripodic Agreement. Indeed, the two documents themselves contemplate each other. See, e.g., doc. 1-2 at 2 (“In the event of any conflict between this Order and the [CSA], the terms of this Order will control.”). While the Order contains no term provision, no choice of law provision, and no warranty disclaimers, the CSA never says what precisely Pruitt bought, nor how much it cost. Only when read as one do the two documents form a complete picture of the parties’ agreement. Reading them together as it must, the Court concludes that the Agreement’s “Initial Term” unambiguously began on June 25, 2013, the day Caradigm and Pruitt executed the CSA and the Order. The Order defines that date as the “order effective date.” Doc. 1-2 at 2. Even more importantly (and understandably, given its designation as the “effective” date), $20,000 monthly payments from Pruitt to Caradigm for a CIP development license began that day. Id. That is, a portion of Pruitt and Caradigm’s respective performances under the Agreement began the same day they executed the Order. As — 1187 — alluded to above, it defies logic to suggest that a contract’s term has not begun even though its performance has.14 The Agreement’s “Testing” provision, which Pruitt believes mandates a First Productive Use start date for the Initial Term, contains nothing on its face that undermines an execution-based beginning. Nor does it suggest a termination avenue outside those included in the termination clause. It contemplates two different types of testing: (1) general testing of “Cloud Services” to ensure conformance with Car-adigm-provided technical guides and user manuals; and (2) testing of “Customer Use Cases,” which are those “developed by or for [Pruitt]” (recall that it paid $20,000/ month for a CIP development license). Doc. 1-1 at 3. General testing ran for a limited time after installation of Cloud Services, while use case testing required only that Pruitt notify Caradigm once testing was “satisfactorily completed” and Pruitt was “ready for Caradigm to move the [use case] to the production environment.” Id. It almost goes without saying that pre-production testing must come before First Productive Use. The Agreement’s “Customer Use Case” testing language confirms that understanding by requiring Pruitt to notify Caradigm “when testing has been satisfactorily completed and [Pruitt] is ready ... to move ... to the production environment,” necessarily implying that notice occurs before production. Doc. 1-1 at 3 (emphasis added). But the fact that testing preceded production in and of itself says nothing about (1) when the Initial Term began, or (2) the termination provision’s applicability during the “testing phase.” Customer Use Case testing also helps confirm that the existence of a testing phase in the parties’ relationship does not inform the Initial Term’s start date. To facilitate that testing, and so Pruitt could develop its own use cases for CIP, Pruitt purchased a development license for $20,000 per month. Doc. 1-2 at 2. In other words, Pruitt performed under the Agreement. As discussed above, once performance under the Agreement has begun, so too must its provisions apply. Nothing about the Agreement generally, or the testing provision in particular, suggests that the termination clause operates differently. Just as testing’s existence plays no role in the Initial Term’s begin date, so too do it and the termination clause fail to interact. Termination pursuant to the Agreement involves no testing related conduct. A party must materially breach (which can occur in countless ways), the other party (the terminating entity) must then give notice, and give the allegedly breaching party ninety days to cure. Doc. 1-1 at 3. None of that necessarily involves testing. Conversely, nothing about testing involves termination-related conduct. The Agreement does not specific precisely what testing involves, but its connection to software performance (whether general or Customer Use Case testing) suggests that notice to the other party (except for use case testing when Pruitt decides its designs work well), an opportunity to cure, and identification of a material breach are not part of the testing equation.15 In other words, the two provisions (testing and termination) govern different conduct and be — 1188 — come applicable and relevant based on different considerations. Pruitt contends that the testing provision’s “satisfactorily completed” language renders its continued participation in CIP implementation contingent, and thus terminable at will prior to testing completion. Doc. 74 at 14-15; see also doc. 67-1 at 16. As Caradigm correctly points out, however, the “satisfaction” language pertains only to Customer Use Case testing. Doc. 71 at 17. It involves a duty by Pruitt—' “will notify Caradigm”—related to software (use cases) that Pruitt purchased a license to create. It would be strange indeed if Pruitt could unilaterally terminate—potentially because of dissatisfaction with something Pruitt created—pursuant to a contract provision that required it, not Caradigm, to act. See, e.g., Clayton McLendon, Inc. v. McCarthy, 125 Ga.App. 76, 79, 186 S.E.2d 452 (1971) (“Where one contracting party agrees to perform services ‘to the satisfaction of or ‘satisfactory to’ the other party, compliance with the contract is not shown unless it appears that the thing done or the article furnished does in fact satisfy the other party.”). Even for those use cases “developed by” Caradigm for Pruitt (doc. 1-1 at 7) though, the notice duty remained Pruitt’s. More importantly, the Agreement’s language does not condition Pruitt’s performance on its satisfaction. Without question use cases cannot move to production until testing results satisfy Pruitt. But nothing in the CSA or Order allows Pruitt to pause development license payments—which it began paying prior to any testing—or otherwise stop performance, if testing leads to dissatisfaction.16 Nothing permits Pruitt to ditch the dispute resolution procedures in the SOW if it becomes dissatisfied. See doc. 1-3 at 27-30 (SOW’s “Project Governance” materials, which outline chains of command and decision making protocols for addressing “issues” and “risks”). And nothing permits Pruitt to avoid First Productive Use (and the increased monthly payments once that occurs) simply because one use case of the four originally ordered failed to satisfy. Instead, dissatisfaction leads to a Pruitt-disapproved use case remaining in pre-production. Doc. 1-1 at 3. Pruitt believes Caradigm’s (and the Court’s) constructions of the Testing provision and termination clause “creates a bizarre contractual framework in which Pruitt could remain lodged in a never-ending implementation—dissatisfied with Caradigm’s solution, incapable of reaching First Productive Use, but unable to pinpoint any uncured ‘material breach’ that would allow it to bring the implementation to a close.” Doc. 74 at 14. Not so. For starters, Pruitt’s dissatisfaction is only relevant to Customer Use Cases. General testing, by contrast, only cares about ,CIP’s conformance to its “technical guides and user manual” and never conditions completion on Pruitt’s satisfaction. Doc. 1-1 at 3, 7. And with use cases, it is entirely conceivable that one could perform poorly and leave Pruitt dissatisfied, while others perform well and thus proceed to production. Under that scenario, the Agreement contemplates First Productive Use beginning, not the across the board interminable pause in CIP implementation that Pruitt fears. Even if a more fundamental problem with CIP (like the inability to match patient data from disparate systems to create the longitudinal patient view Caradigm marketed) undermined all use cases and delayed First Productive Use, the nightmare scenario Pruitt envisions need not necessarily come to pass. — 1189 — Imagine the consequences of Pruitt’s hypothesized “endless loop of contractual purgatory.” Doc. 67-1 at 18. Prior to First Productive Use, it does not like CIP’s performance as to any use cases, but cannot identify a true Caradigm material breach. What to do? Under the Court’s construction of the termination provision, and absent a subjective satisfaction requirement, Pruitt cannot unilaterally walk away. That, however, is not the same thing as saying Pruitt is permanently stuck in an unhappy marriage. Pruitt and Caradigm could always mutually agree to walk away. What’s more, the SOW provides detailed provisions governing processes by which the parties would work out disputes, including those about performance. Pruitt could have availed itself of those rather than give up. It did not. Pruitt also could have pursued a damages claim regarding any non-material breach it thought Cara-digm committed. It did not. At bottom, Pruitt had options, all of which remained unexercised prior to its unilateral termination, other than the contractual purgatory it imagined. Nothing about either type of testing (general or “Customer Use Case”), read in isolation or a larger context, allows Pruitt to unilaterally terminate if dissatisfied. Instead, they simply highlight a common sense feature (if not one as spelled out as one might hope) of the Agreement—a pre-First Productive Use phase that involved everything from installation to testing that the parties needed to ensure that CIP functioned in a way that would benefit the customer (Pruitt) rather than weigh down its operational efficiency. See generally doc. 1-3. Finally, both parties cite to evidence outside the four corners of the Agreement to support their respective constructions. See, e.g., doc. 80 at 5 (Pruitt pointing to Caradigm’s CEO’s letter response to Pruitt’s termination to support its contention that the “Initial Term” ran for five years). But such parol evidence cannot be considered where no ambiguity infects the Agreement. See Brazeal v. Newpoint Media Grp., LLC, 331 Ga.App. 49, 57, 769 S.E.2d 763 (2015) (“[WJhere an action is brought upon an unambiguous contract, parol evidence may not be introduced to explain the intention of the parties in entering into the agreement.”). None does regarding the Initial Term, testing, or the termination clause’s applicability. Any evidence of the parties’ respective understandings of those terms, is, at best, evidence of their mistake of law and not relevant to contract construction. To be clear: the Court finds that the Agreement unambiguously provides that the Initial Term began the day the parties executed the Agreement, not whenever First Productive Use occurred. Further, the Agreement’s Testing provision says nothing about termination. In that vein, it does not provide that Pruitt may walk away if it grows dissatisfied with CIP or Caradigm. Termination is instead governed exclusively by the termination clause.17 Because that allows for unilateral — 1190 — termination only'after notice to the allegedly breaching party and an opportunity to cure within ninety days, Pruitt’s decision to walk away from CIP itself amounted to a breach unless it satisfied those requirements. ii. Pruitt’s (Non-)CompIiance with Notice and Cure Under Georgia law ‘[t]he general rule in determining contract compliance is substantial compliance, not strict compliance.” Del Lago Ventures, Inc. v. QuikTrip Corp., 330 Ga.App. 138, 142, 764 S.E.2d 595, 598 (2014) (quoting Rome Healthcare LLC v. Peach Healthcare Sys., Inc., 264 Ga.App. 265, 272, 590 S.E.2d 235, 241 (2003)); see also O.C.G.A. § 13-4-20 (West 2010). Substantial compliance is also the general rule for notice provisions. See Del Lago, 330 Ga.App. at 144, 764 S.E.2d at 599. However, substantial compliance is not the rule when a contract clearly establishes written notice as a condition precedent to suit. See Pillar Dev., Inc. v. Fuqua Constr. Co., Inc., 284 Ga.App. 858, 860, 645 S.E.2d 64, 66 (2007) (“Where a contract contains a provision requiring written notice of a claim for breach, ‘[t]he failure to give notice as required or to show waiver by [the party entitled to notice] is an independent bar to the maintenance of a successful cause of action on the contract.’ ” (quoting Orkin Exterminating Co. v. Stevens, 130 Ga.App. 363, 369, 203 S.E.2d 587, 593 (1973)) (alterations in original)). Critically, “oral notice is not sufficient where written notice is required.” Id. at 860, 645 S.E.2d at 66; see also Eells v. State Farm Mut. Auto. Ins. Co., 324 Ga.App. 901, 904, 752 S.E.2d 70, 73 (2013) (holding that oral notice did not comply with an insurance contract’s requirement for written notice of a claim); Moss v. Cincinnati Ins. Co., 154 Ga.App. 165, 165, 268 S.E.2d 676, 677 (1980) (“Even if appellants established that oral notice had been given, it would not satisfy the written notice requirement.”). Triad Constr. Co. v. Robert Half Int’l, Inc., No. 16-14942, 679 Fed.Appx. 748, 752, 2017 WL 491154, at *3 (11th Cir. Feb. 7, 2017). The Agreement allows either party to terminate “if the other party commits a material breach ... that remains uncured for 90 days after written notice of such breach.” Doc. 1-1 at 3. It never defines what constitutes notice beyond the mandate that it be written, though it does spell out where to send notice and how to deliver it. Id. at 5. Pruitt believes that it “substantially complied with [the] termination provision’s notice requirement,” either by combining a January 29, 2015 phone call with a February 17 email, or via its letter sent approximately one month after Caradigm filed suit. Doc. 74 at 15-16. The phone/ email combination fails to qualify because Pruitt never gave Caradigm the opportunity to cure. Instead, it said, at best, the equivalent of “we are walking away from this contract, and here’s why.” Nothing about “we would like you to fix x, y, and z,” nothing even implying that Pruitt would welcome a Caradigm attempt to fix the problem, and nothing referencing the Agreement-mandated ninety day cure period. What’s more, neither the phone call, the email, nor a combination of the two gave notice of a material breach, much less for purposes of giving Caradigm a chance to cure. Instead, as Caradigm correctly observes, both served as notice of Pruitt’s termination of the Agreement. That jumped the gun and cannot qualify as notice that satisfies the termination clause. See Woodall v. Pharr, 119 Ga.App. 692, 694, 168 S.E.2d 645 (1969) (purported no — 1191 — tice that “precluded correction of the default ... was not sufficient to constitute the notice of default required” by a lease). The post-suit letter also falls short because Pruitt sent it after Caradigm filed suit. Regardless, the letter never offers Caradigm an opportunity to cure. Instead, it serves “as written notice of material breach,” and asks that Pruitt not “host any [Pruitt] data.” The Agreement requires more. It mandates that (1) the terminating party give written notice of material breach (check), and (2) wait ninety days for the breaching party to cure before allowing a walk-away. Doc. 1-1 at 3. Pruitt’s letter checked the first box, but not the second. Like the email-phone call combo, it too fails to satisfy the termination clause. Pruitt’s proposed construction is a reasonable way to structure a contract. But it is not how these parties structured this contract. For better or worse, the termination clause “went live” the day Pruitt and Caradigm executed the Agreement. Because (1) that clause provided the only way for a party to unilaterally terminate, (2) Pruitt failed to comply, Pruitt’s decision to walk away breached the Agreement. See Flint Emergency Med., LLC v. Macon Cty. Med. Ctr., Inc., No. 5:12-CV-105-MTT, 2013 WL 5507279, at *8 (M.D. Ga. Oct. 2, 2013). (“[A] party that terminates the Agreement without [giving notice and an opportunity to cure] breaches the Agreement with respect to the manner of its termination, but that party does not preclude itself from asserting other claims for breach pursuant to the Agreement. Nor has it undercut its ability to raise its opponent’s breaches as a defense. It simply has opened itself to liability for breaching [the Agreement] through an improper termination.”). iii. Anticipatory Repudiation In addition to its termination clause breach theory, Caradigm argues that Pruitt anticipatorily repudiated, and thus breached, the Agreement. “ ‘[A]ntici-patory repudiation’ of a contract occurs when one party thereto repudiates his contractual obligation to perform prior to the time such performance is required under the contract.’” John K. Larkins, Jr., Ga. Contracts Law and Litigation § 11:5 (2d ed.) (quoting CCE Fed. Credit Union v. Chesser, 150 Ga.App. 328, 330, 258 S.E.2d 2 (1979)); see also Clark v. Cox, 179 Ga.App. 437, 437, 347 S.E.2d 4 (1986) (“When appellant repudiated his contractual obligation to perform prior to the time performance was required, an anticipatory repudiation of the contract occurred, forming the basis of appellee’s breach of contract action.”). The party must “absolutely” refuse to perform, and that refusal must be unqualified. Textile Rubber & Chem. Co. v. Thermo-Flex Techs., Inc., 301 Ga.App. 491, 494, 687 S.E.2d 919 (2009). In Clark, for example, a real estate buyer inspected the property the day before closing and noticed several conditions that would, if present at delivery, constitute breaches of the parties’ contract. 179 Ga.App. at 437, 347 S.E.2d 4. He declared his intention not to close (ie., not to perform) since “there was no way that the items could be remedied” before closing. Id. The next morning, the buyer “did not appear at the scheduled closing.” Id. That, held Clark, qualified as contract repudiation “prior to the time performance was required.” Id. And that “absolved the [seller] from remedying the situations of which [the buyer] complained.” Id. Clark therefore affirmed summary judgment in the seller’s favor.- Like the Clark buyer, Pruitt stopped performing by unilaterally “suspending] all efforts dedicated to [the CIP] project,” including “utilization of [Cara-digm’s] services. Therefore,” said Pruitt at the time, the Agreement “[was] no longer — 1192 — valid.”18 Doc. 55-26 at 2. That is the quintessence of repudiation. Pruitt “absolutely” refused to perform and did so without qualification. See Textile Rubber, 301 Ga.App. at 494, 687 S.E.2d 919. But did that repudiation come before the time for Caradigm’s patient matching performance? Yes. When Pruitt terminated the Agreement, the parties had not held the contemplated “kick off’ meeting that marked the transition from “pre project” to “project start” in the SOW. Doc. 73 at 24-26 (Pruitt’s Reponse to Caradigm’s Statement of Undisputed Fact); doc. 1-3 at 21 (SOW’s “Indicative Deployment Schedule & Process”). They never built the data feeds, much less actual use cases. Doc. 73 at 8. Per the SOW, the “project” had yet to really begin. See doc. 1-3 at 21.19 At best, CIP’s implementation remained in its early stages, and nowhere near First Productive Use, at the time that Pruitt stepped back. The Agreement never specifies precisely when Caradigm had to provide “go live” quality patient matching. But it is clear that First Productive Use—which relied on functional use cases, which in turn required adequate patient matching—would not occur until use case performance satis,fied Pruitt. Doc. 1-1 at 3. It’s also clear that the approximately six to seven month “ramp up” period immediately prior to First Productive Use would have required accurate patient matching much more so than the preliminary phases the project actually reached when Pruitt terminated. At that time, the parties were still setting up data feeds. In some sense, then, CIP remained partially inoperational (much less testable, much less able to integrate seamlessly into Pruitt’s cornucopia of IT systems) when Pruitt abandoned the ship. Regardless of precisely, when Pruitt’s patient matching performance needed to function at a “go live” level, the Court has no trouble concluding that it need not have done so at the early stage at which Pruitt terminated. Pruitt insists that Caradigm had already failed to perform patient matching when it terminated the Agreement. Doc. 74 at 20. When Pruitt delivered a.small batch of data with which to test patient matching, Caradigm returned what Pruitt considered an unacceptably low match rate. Doc. 74 at 20-21. Caradigm’s failures, in Pruitt’s eyes, “were real, not speculative.” Id. at 21; id. (“Pruitt’s dissatisfaction was not with how Caradigm might perform; it was how Caradigm actually performed when given an opportunity.”) (emphasis in origi — 1193 — nal). And that, says Pruitt, precludes Cara-digm’s quest for damages. Id. at 19 (citing Beauchamp v. Coastal Boat Storage, LLC, 4 So.3d 443, 451-52 (Ala. 2008)). Even taking Pruitt’s word that the patient matching test run returned “abysmal” results, those occurred before the time at which Caradigm had to provide “go live” quality matching capabilities. Cara-digm may have inspired little confidence when it whiffed at its first attempt to impress. But the Agreement did not consider that swing and a miss, at that stage, a material breach. Indeed, it anticipated hiccups and provided procedures and governance provisions to address those situations. What’s more, the Agreement’s warranty and medical diagnosis disclaimers suggest that Caradigm’s patient matching abilities (or lack thereof) did not qualify as nonperformance. Patient matching appears only twice in the Agreement. One comes in an all upper-case “Disclaimer of Warranties” provision. Doc. 1-1 at 5 (“CARA-DIGM PROVIDES NO OTHER EXPRESS OR IMPLIED WARRANTIES OR CONDITIONS WITH RESPECT TO SOFTWARE OR SERVICES ... INCLUDING PATIENT RECORD MATCHING.”). The other makes clear that “false matches” and “errors or omissions in the clinical information” may occur in matched records “[bjecause patient information is maintained in multiple places ... not all patient information is kept in a standard fashion or is regularly updated,” and because of the “statistical nature of the algorithm used to match patient identities.” Id. at 4. The Agreement thus advised that Pruitt “verify clinical information with each patient ... before such information is relied upon or used.” Id. Pruitt likens those warnings to a car maker advising that “a car may experience mechanical issues.” Doc. 74 at 21. That type of warning, says Pruitt, “does not mean that a manufacturer can claim it sufficiently performed if the engine only runs half the time.” Id. Maybe so. But if the contract between the car maker and a dealer expressly disclaimed any warranty as to engine performance, the dealer would have a hard time recovering for a failure to perform based on the engine not running. What matters is what the Agreement says. And the Agreement says that Cara-digm did not warrant its patient matching in any way. It also warned Pruitt not to rely on matched records without verifying their accuracy. The only takeaway from those unambiguous provisions is that Pruitt had no contractual right to complain about a particular patient matching performance level. Less than ideal matching could have delayed .the project, but at such a preliminary stage of the relationship it fell short of a material breach. Despite no right to complain about something that Caradigm had no obligation at the time to perfect, Pruitt unilaterally terminated the Agreement because of Caradigm’s perceived inability to patient match. Put differently, Pruitt “repudiated [its] contractual obligation to perform pri- or to the time [Caradigm’s] performance was required.” Clark, 179 Ga.App. at 437, 347 S.E.2d 4. In doing so, it breached the Agreement. The Court accordingly GRANTS Cara-digm’s summary judgment motion and DENIES Pruitt’s as to the parties’ breach claims. The matter of damages remains. 2. Damages Caradigm contends that the Agreement “requires summary judgment in favor of Caradigm on its contractual claim for interest, costs and fees.” Doc. 65-1 at 26. In fact, says Caradigm, Pruitt’s “anticipatory breach and improper termination made it liable to pay ... Caradigm’s full benefit of the bargain.” Id. On top of that, Caradigm argues that Pruitt owes it “1.5% monthly — 1194 — interest” pursuant to the Agreement’s payment provision from the date it filed suit. Id. Pruitt sees Caradigm’s damages analysis as fatally flawed. Even if the Court finds Pruitt liable for breach it cannot, Pruitt says, be made to pay “the totality of profit Caradigm would have received in a successful implementation plus fees and interest on that accelerated amount at 1.5% per month.” Doc. 74 at 22 (emphasis in original). That “entire” value of the Agreement, says Pruitt, “was not ascertainable at the time of the alleged breach.” Id. at 23. Since “Georgia law does not allow for the ‘acceleration’ of monthly payments of a speculative and indefinite duration,” Pruitt cannot be made to pay that “entire” amount. Id at 23. Even if it could, the CSA’s 1.5% interest provision only applies to “late payments” or “past due amounts,” not to “fees [Caradigm] was to receive, if at all, only many years in the future.” Id at 24. First, Caradigm is not entitled to 1.5% interest on anything (whether it may seek statutory interest is another, unargued, question). The Agreement provides that “[a]ny late amounts” accrue interest monthly at that rate. Doc. 1-1 at 3. Those “amounts,” however, become “late” only if not paid “within 30 days after the invoice date.” Id. Pruitt timely paid all amounts for which Caradigm sent it an invoice. It certainly has not paid monthly licensing fees or any other amounts contemplated by the Order since it terminated, but then again Cara-digm never invoiced those fees. Caradigm thus cannot seek interest for payments it lost because of Pruitt’s early termination. The usual meaning of the term “late amount” reinforces that conclusion. If a tenant signs a one-year lease, pays for two months, then leaves unexpectedly, he or she is almost certainly on the hook for the remaining ten months of lease payments. But any late fee provision of the lease will not apply. The same is true here. Regardless of what amounts Pruitt may owe, it never submitted a “late payment” as that phrase is commonly understood. Interest aside, Caradigm’s claim to the “entire value” of the Agreement also holds little water. It’s true as a general proposition that an anticipatory repudiation “accelerates the maturity of [an] indebtedness.” Doc. 78 at 14 (quoting 57 Causes of Action 2d 587 (2013)); see also Parker v. King, 68 Ga.App. 672, 23 S.E.2d 575, 577 (Ga. Ct. App. 1942) (a party injured by an anticipatory breach “may at his election at once sue and recover his entire damages”). But what is the entire value of the Agreement? How can the Court measure that value? The Agreement is not a note, bond, or contract for sale of a good or service with a fixed value. See, e.g., Textile Rubber & Chem. Co. v. Thermo-Flex Techs., Inc., 308 Ga.App. 89, 89, 706 S.E.2d 728 (2011) (defendant’s anticipatory repudiation of $3 million contract for sale of carpet-making technology allowed plaintiff “to issue a demand for the face value of the total remaining unpaid debt”). Anticipatory repudiators of those contracts understandably incur liability for the their entire value. Here, however, the Agreement’s entire value “was,” as Pruitt notes, “not ascertainable at the time of the alleged breach.” Doc. 74 at 23. Assume Pruitt never walked away. If pre-First Productive had lasted six months, the “entire value” would be one amount. If it lasted ten, another. Then again, it might have been the ease that Caradigm never improved its patient matching and CIP implementation never occurred. Nevertheless, all those arguments pertain to the amount of the Agreement’s value, not whether Caradigm as the prevailing party may recover that value. Geor — 1195 — gia law is clear: the Agreement’s entire value became due the moment Pruitt breached. Normal damages principles (such as the bar on speculative recoveries) will, of course, apply at a trial to determine how much Pruitt owes. But none of those derail the Court’s liability conclusion that leads to the entire Agreement value becoming due. IV. CONCLUSION Accordingly, the Court GRANTS Cara-digm’s motion for summary judgment (doc. 65), and DENIES Pruitt’s. Doc. 67. Cara-digm’s motion to exclude Cheri Kane’s testimony is DENIED AS MOOT. Doc. 75. The parties must file a proposed pretrial order within thirty days of the date this Order is served. SO ORDERED, this 30th day of May 2017. Footnotes [1] All citations are to the electronic docket unless otherwise noted, and all page numbers are those imprinted by the Court’s docketing software. [2] The CSA defines an "order” as “a document signed by both parties that specifies Client Software, Third Party Software, Cloud Services, and/or other Services to be licensed or purchased by Customer under” the CSA. Doc. 1-1 at 7. [3] The SOW, which focuses intently on pre-productive use phases of the deal, actually divides the CIP "deployment schedule and process” into three phases - (1) pre-project; (2) "foundations + use case;” and (3) "go live.” Doc. 1-3 at 21. Phases one and two are divided by a "project start,” signaled by the kick off meeting. Id.; doc. 65-1 at 6. The project transitions from phase two to three when Pruitt engages in a First Productive Use. Doc. 1-3 at 21. [4] The CSA defines "Use Case” as a "Cloud Services configuration that incorporates Data Feeds with designated views to address specific issues within [Pruitt] or [Pruitt's] Facilities." Doc. 1-1 at 7. "Customer Use Cases,” in turn, "means any Use Case developed by or for [Pruitt] or customized for [Pruitt].” Id. [5] "Data Feeds” are "inbound electronic data .messages from [Pruitt] information systems to the Cloud Services.” Doc. 1-1 at 7. Essentially, a data feed is the delivery of information from a Pruitt electronic medical records software system to CIP, which Caradigm hosts on its own computers, and which Pruitt accesses via the "cloud.” See doc. 73 at 6 ("data feeds are the electronic data messages from Pruit-tHealth systems to [CIP].”) (quotes omitted). [6] The date for the "kick-off" remained undefined because of the inherent uncertainty surrounding how long it would take to complete necessary preliminary tasks. [7] The CSA contains a variety of other provisions beyond those that discuss contract term, termination, testing, and warranties. Important under circumstances not presented by this dispute, those provisions have no relevance to the claims and defenses the parties present. [8] Beyond warranties, the CSA warned that Pruitt [was] solely responsible for use of the Cloud Services and for verifying the accuracy and adequacy of information and data furnished for processing.... Because patient information is maintained in multiple places, not all of which are accessible, not all patient information is kept in a standard fashion or is regularly updated and, because of the statistical nature of the algorithm used to match patient identities with their clinical information, it is possible that false matches may occur or that there may be errors or omissions in the clinical information. Doc. 1-1 at 4. [9] The Court recognizes that briefs are not undisputed facts and thus cannot undergird summary judgment findings. The "patient matching” quote above serves no such purpose. Nevertheless, the quote, from a portion of Pruitt’s opening summary judgment brief, is the most succinct and accurate description of the term “patient matching” in the record. The Court uses it for that illustrative purpose only. See also doc. 75 at 5 (Caradigm’s opening motion to exclude brief) ("Patient matching is the act of matching a patient's records across multiple electronic medical record sys- . terns.”) (quotes omitted). [10] If this case proceeds to a damages trial, Caradigm may renew its motion should Pruitt seek to use Kane's testimony at that point. [11] In other words, if the Court agrees with Caradigm’s view of the Agreement, its claims succeed and Pruitt's necessarily fail as a matter of law. [12] The CSA requires that "the laws of the state where [Pruitt’s] headquarters are located” govern the construction of its terms. Doc. 1-1 at 5. Pruitt’s headquarters are in Georgia (doc. 7 at 22), hence Georgia law applies. [13] That’s not entirely true. The Agreement allows a party to "terminate an Order by providing written notice of non-renewal at least 90 days before the end of the then-current Term” (either Initial or a Renewal Term). Doc. 1-2 at 3. And terminating the Order—i.e., one for CIP services—also terminates the Agreement itself. IA But neither party argues that Pruitt attempted to utilize that off ramp, so its mention would just muddy the analytical water. [14] Contracting parties could avoid the "performance equals term start" conclusion by explicitly allowing that some performance will occur prior to a term (or termination) provision's application. That did not happen here. [15] Caradigm could conceivably do something during testing that qualifies as a material breach that itself unlocks the termination clause. But the conduct that the testing provision contemplates does not inherently involve, for example, shoddy work. [16] In fact, "[f|ailure to make timely payment is a material breach of” the Agreement. Doc. 1-1 at 3. [17] Pruitt’s insistence that ”[n]othing in [the Termination provision] provides that an uncured material breach is the exclusive mechanism by which either party can terminate implementation,” and that Caradigm "myopically” focuses on that provision, inverts the significance it plays in a proper reading of the Agreement. True, the Termination provision never says "this is the only way to end the Agreement.” But it did not need to. The mere existence of a provision devoted solely to the Agreement’s duration and how to end it necessarily implies, absent explicit text to the contrary elsewhere in the contract, that it contains the sole means by which a party may unilaterally terminate. It thus makes sense when arguing that one party breached by walking away to "myopically” focus on that provision. [18] Pruitt’s performance took many forms. It paid Caradigm a monthly licensing fee, for example. It also promised to staff CIP implementation with a number of employees serving in specific roles, "[b]e involved in all aspects of the Project,” and "[cjomplete the responsibilities identified in th[e] SOW in a timely and effective manner,” among other things. See, e.g., doc. 1-3 at 22, 24. Although Pruitt never failed to make a monthly payment for months during which it participated in the project (and apparently made at least one payment after it unilaterally withdrew), its February 2015 walk away ended its performance in countless other ways. [19] Pruitt objects to Caradigm relying on’ the SOW’s project timeline. See doc. 74 at 3 ("The illustrative chart in the SOW is not contractually significant.”). Although that timeline indeed reflects only "an estimate of the phases and milestones for performance of the Project” (doc. 1-3 at 21), the SOW serves as part of the overall agreement. See doc. 1-1 at 2; doc. 7 at 30 ("Prior to their termination, the [CSA], as amended, the ... Order, and the SOW were valid and enforceable agreements.”). It thus best reflects the parties’ intentions regarding their CIP-related relationship. See Greenberg Farrow Architecture, Inc. v. JMLS 1422, LLC, 339 Ga.App. 325, 329, 791 S.E.2d 635 (2016) (to ascertain the parties’ "true intent,” courts look to unambiguous contract terms).