This post uses hypothetical scenarios for illustrative purposes only. It does not describe any actual client, transaction, or representation, and is not legal advice.
Here is a pattern that shows up in almost every signed-but-not-closed deal: the buyer keeps running its business. It answers emails. It makes ordinary commercial decisions about partners, integrations, pricing, and roadmaps. Nobody on the deal team is thinking about the termination section of the merger agreement when a mid-level executive declines a competitor’s request for a custom integration. That is just Tuesday.
On August 7, 2026, Vice Chancellor Bonnie W. David issued a post-trial opinion in Verisk Analytics, Inc. v. ExactLogix, Inc. d/b/a AccuLynx.com, C.A. No. 2026-0023-BWD, holding that exactly that kind of ordinary Tuesday decision stripped a buyer of its right to walk away from a $2.35 billion acquisition — and ordered it to close. The clause that did the work was four words long. Most of the lawyers reading this have signed agreements containing those four words without a second thought.
The deal, the email, and the outside date
Verisk agreed on July 29, 2025 to acquire AccuLynx, a roofing-contractor software platform, for $2.35 billion in cash through a reverse triangular merger. Verisk’s own product line included Xactimate, the property-damage estimation tool used across the insurance claims industry. Before signing, Verisk had been in discussions with ServiceTitan — an AccuLynx competitor — about a deeper, bespoke integration with Xactimate pricing data.
Roughly five days before signing, Verisk decided internally to pause those discussions and offer ServiceTitan only the baseline integration available to everyone. Seven days after signing, on August 5, 2025, Verisk sent the email implementing that decision. The email told ServiceTitan that now that the AccuLynx acquisition had been announced publicly, and in light of new information, Verisk would not be able to continue down the path of placing Xactimate pricing data directly into ServiceTitan’s estimating solution.
ServiceTitan took that email to the FTC. What followed consumed the deal’s entire runway. The FTC opened a preliminary investigation, developed a “market reset” theory of harm, and on October 22, 2025 issued a Second Request. Verisk collected nearly four million documents from sixteen custodians, produced over 400,000, and spent roughly $8 million on outside counsel and document vendors. On December 24, 2025, the FTC confirmed it would require full compliance. Two days later Verisk terminated.
The outside date was the problem. The parties had negotiated a termination date of November 26, 2025 — four months after signing — with one one-month extension, which Verisk exercised. AccuLynx had originally proposed ninety days; Verisk had proposed six months. Nobody built in an automatic extension triggered by a Second Request. A single Second Request in an HSR-reportable multibillion-dollar deal ate the whole window, which is roughly what anyone who has lived through second-request compliance would predict.
The four words that decided the case
Section 9.1(f) gave Verisk the outside-date walk right, subject to a proviso that it could not terminate if a buyer party’s “knowing and willful breach” had prevented satisfaction of the closing conditions. That is the familiar formulation, and Verisk was comfortable with it — the court found no evidence that Verisk had intentionally sabotaged the FTC process, and AccuLynx did not even argue that prong.
But the final sentence of Section 9.1 said something broader. Termination was unavailable to a party if that party’s failure to fulfill its obligations or comply with its covenants, or other willful conduct, had been the primary cause of, or primarily resulted in, the failure to satisfy any closing condition.
Verisk argued “willful conduct” had to mean something in the neighborhood of a breach. The court rejected that, noting Verisk could not explain what it means to be similar to but not quite a breach. Instead the court read the agreement’s own internal vocabulary against Verisk. Section 9.2 carefully defined “Willful Breach” as an intentional and willful material breach undertaken with the intent of causing a breach. Section 9.1(f) used “knowing and willful breach.” And then the very next sentence used bare, undefined “willful conduct.” Juxtaposing those terms, the court held, ratchets the required mental state down. “Willful” was left to mean simply voluntary and intentional — no wrongfulness, no blameworthiness, no bad purpose, and critically, no breach required at all.
That is the whole case. Declining to build a custom integration for a competitor was lawful. It was commercially rational. Verisk did not expect it would raise competition concerns. The court said none of that mattered: terminating the discussions was voluntary and intentional conduct even if doing so was not wrongful. Ordinary-course business judgment became a termination-blocking trigger.
Causation was the harder half, and the buyer’s own expert lost it
The final sentence required more than mere contribution. The court read “primary cause of, or primarily resulted in” as a but-for and proximate-cause standard — and expressly noted it sets a higher bar than the common-law prevention doctrine, under which a breach need only have contributed materially to the non-occurrence of a condition. So the parties had, perhaps without noticing, turned two dials in opposite directions: down on mental state, up on causation.
AccuLynx cleared the higher bar anyway. The most damaging testimony came from Verisk’s own regulatory expert, a former Director of the FTC’s Bureau of Competition, who testified that the August 5 email was the reason the FTC needed to investigate. The court also noted that ServiceTitan was the only AccuLynx competitor with numerous calls with the FTC, the only one subpoenaed, and the only one the agency engaged with on critical dates. More likely than not, the court concluded, the FTC would not have been seriously concerned about a market reset but for that email.
Notably, the court declined to decide whether Verisk breached its commercially reasonable efforts covenants at all, calling that theory an awkward fit for these facts and observing that AccuLynx’s position largely rested on the questionable premise that sophisticated parties just need to do better when they respond to regulatory authorities. That is a useful data point for anyone who assumes an efforts clause is the natural home for a regulatory-conduct dispute. Here, it was not the efforts covenant that bound the buyer — it was a sentence most negotiators treat as boilerplate. We have written before about how divestiture caps and hell-or-high-water obligations allocate regulatory risk on the front end; this opinion is a reminder that the back end of the agreement can quietly override the allocation you thought you negotiated.
Specific performance, because the parties said so
Having found the termination invalid, the court granted specific performance and ordered Verisk to use commercially reasonable efforts to obtain HSR clearance and close if the FTC approves. Verisk’s argument that damages would be adequate went nowhere, because Section 11.12 contained the standard irreparable-harm stipulation in which each party agreed not to raise any objection to the availability of specific performance. Delaware courts enforce that bargain. As the court put it: “It is not inequitable to hold the parties to their bargain.”
One detail deserves attention from sell-side lawyers. After the purported termination, AccuLynx ran a process — its banker approached at least ten potential buyers and generated three indications of interest, one contemplating a revised valuation and signing within two weeks. That is a straightforward breach of the exclusivity provision, and it did not defeat specific performance. Why? Because Verisk never argued the breach was material, and the court found it immaterial since it followed the purported termination and did not deprive Verisk of its bargain. The lesson runs both directions: buyers should deem no-shop breaches material and a bar to equitable relief, and litigants must actually brief materiality. The court also awarded AccuLynx $3.85 million in direct costs — in part because Verisk simply failed to respond to the request in its answering brief and thereby waived any opposition.
What to change in the next agreement you paper
First, treat any termination blocker keyed to “conduct” rather than “breach” as a red-flag term. If the parties want to foreclose termination only where wrongful conduct caused the condition failure, they can use the defined term. Delete “or other willful conduct,” or tie it expressly to the defined “Willful Breach,” or require both knowledge and intent to cause the breach.
Second, understand that carefully defining a term in one section creates a negative inference against every undefined cousin nearby. In XRI Investment Holdings LLC v. Holifield the same interpretive move ratcheted a standard up; here it ratcheted down. If you define “Willful Breach,” either use it everywhere or state expressly that related terms carry the same meaning.
Third, mind the signing line. Verisk argued it could not have breached before the agreement existed, and it was right about the decision — which was made five days pre-signing. But it implemented that decision seven days post-signing. A buyer with a known pre-signing decision should either execute it before signing, disclose it on the schedules, or negotiate a carve-out for conduct decided upon or disclosed prior to signing.
Fourth, size the outside date to the regulatory reality and price the walk right. Four months plus one is not a Second Request timeline. Consider an automatic extension on issuance of a Second Request, extensions keyed to substantial-compliance certification, and — if the real fear is the duration and cost of review rather than an outright block — a reverse termination fee or ticking fee rather than an outside date that functions as the only exit. Verisk’s $8 million and its 18-to-24-month compliance estimate bought it nothing.
Fifth, consider an interim covenant that actually addresses the risk. Nothing in the quoted covenants barred Verisk from taking business actions likely to invite antitrust scrutiny of the very deal it had signed, and nothing required that regulatory submissions be complete and accurate or that a reasonable internal search precede a response. A seller-side covenant prohibiting conduct reasonably likely to delay or impair clearance would have turned a strained efforts argument into a clean one. Buyers, of course, should resist exactly that.
Sixth — and this one is free — nobody between signing and closing should put “because of the acquisition” in writing when declining a competitor’s request. The August 5 email did the agency’s work for it. Deal teams that understand what is and is not permissible between signing and closing tend to route these communications through counsel, which is cheap insurance compared to a second request.
None of this is unique to billion-dollar software deals. Florida middle-market purchase agreements are papered off the same forms, with the same final-sentence termination blocker and the same Section 11.12 specific-performance stipulation, often with far less negotiation. A buyer that assumes an outside date is a reliable exit — and that its exposure to being forced to close is theoretical — should read this opinion carefully. The drafting cannot guarantee a clean walk right, but the difference between four words present and four words absent is the difference between a $2.35 billion exit and a court order to buy the company.
The full opinion is available from the Delaware Court of Chancery.
If you are negotiating termination rights, outside dates, or regulatory covenants in a purchase agreement, feel free to reach out to our firm manager, Magda, at Magda@montague.law, or fill out our contact form. Mention you read this post.
