The Jira Tickets That Cost Verisk $2.35 Billion

A Delaware judge didn't decide whether Verisk should buy a roofing-software company. The judge looked at Verisk's own project-management database and decided that Verisk hadn't tried hard enough to get the deal through regulators, so it had to stay.

The weirdest part of the Verisk-AccuLynx saga isn't the $2.35 billion in cash price tag, the FTC investigation, or even the fact that a buyer got ordered to close a deal it no longer wanted. The weirdest part is that the smoking gun was a Jira database.

Judge Bonnie David of the Delaware Chancery Court reviewed ordinary-course data from Verisk's Xactware division, maintained in a Jira project-management system, spanning from January 1, 2023, to October 14, 2025. Those internal work records showed the judge that Verisk had engaged in "willful conduct" that caused a regulatory closing condition to fail. The company hadn't earned the contractual right to walk away. So on August 7, the judge ordered Verisk to consummate the merger and told AccuLynx it was entitled to recover its direct costs plus interest.

That was not the story Verisk wanted to tell.

Here's the machinery. Verisk Analytics — a data company embedded in the U.S. insurance claims ecosystem, running products like Xactware that estimators use to price damage — signed a definitive agreement in July 2025 to acquire AccuLynx for $2.35 billion in cash. AccuLynx is the dominant SaaS platform for residential roofing contractors, most of whom perform insurance-driven repairs. The deal was supposed to close by the end of the third quarter of 2025. It didn't make it through the summer.

Instead, the Federal Trade Commission extended its review, issued a "second request" — the formal term for an in-depth antitrust investigation, usually a sign the agency is seriously considering a challenge — and by late November was still asking questions about "non-horizontal" concerns. That phrase sounds technical but it just means the FTC wasn't worried about Verisk and AccuLynx competing for the same customers. It was worried about something else: probably data network effects, or the way combining claims estimation with the platform that contractors use to manage repair workflows could give Verisk unusual leverage across the insurance restoration pipeline.

The contract had a termination date of December 26, 2025. The FTC hadn't cleared the deal. On December 29, Verisk announced it was pulling the plug. The company also said it would redeem $1.5 billion in senior notes... at 101% of principal plus accrued and unpaid interest. The notes contained what the press release called a "special mandatory redemption provision" — which is the debt-market way of saying the bonds had to come back if the deal didn't close.

AccuLynx immediately declared the termination invalid. Verisk said it disagreed and would "vigorously defend" its position. In January 2026, Verisk sued in Delaware's Chancery Court, asking a judge to confirm its exit was lawful under the contract.

Here's where the story stops being about the FTC and starts being about what M&A contracts actually say versus what people think they say.

Most merger agreements include a condition that the deal must receive regulatory approval before it closes. They also include a termination right that lets the buyer walk away if that condition isn't met by a certain date. But there's a standard clause sitting right next to both of those provisions, and it's the one Verisk ran into. It requires both parties to use "commercially reasonable efforts" to obtain the necessary regulatory approvals.

The basic point is that you can't just wait for the clock to expire and then claim the condition failed. You have to actually try to get the deal through. The clause exists because regulators are often slower on deals the companies themselves don't care about. If one party wants out, it has a perverse incentive to do the bare minimum on its regulatory filings and then blame the agency when the deadline passes.

Delaware judges have been enforcing this requirement with increasing sharpness. In this case, the judge had the buyer's own project tickets. The finding of "willful conduct" means Verisk's internal records showed it hadn't been acting in good faith to get the deal cleared. The termination right didn't trigger. And when a Delaware equity court decides a party breached the duty to use commercially reasonable efforts, it doesn't just say the termination was improper. It can order specific performance — the court's equivalent of: you must close the deal.

The funny thing, if "funny" means structurally revealing, is how quickly Verisk went from acquirer to hostage. The $1.5 billion in acquisition-linked debt already has to be redeemed at 101 percent. Now the company also needs to find $2.35 billion in cash to buy a roofing-software platform it apparently no longer wanted. That's not just a strategic reversal — it's a mechanical one. The money raised to fund this deal is being sent back to bondholders, and now fresh money has to come from somewhere else.

As of September 30, 2025, Verisk's leverage would have been 1.9 times trailing adjusted EBITDA pro forma for the debt redemption. The company had $1.2 billion in remaining share repurchase capacity. VRSK trades at roughly $192 today, down about 14 percent year-to-date and roughly 28 percent on a rolling annual basis. The company isn't exactly sitting on a war chest it can deploy without consequences.

The simplest model here is a lesson in the plumbing of M&A exit rights. The buyer thinks: the regulatory condition wasn't met, so I'm out. The contract actually says: the regulatory condition wasn't met, so I'm out — but only if I can prove I was actually trying to get it met. And if the buyer's own project tickets, internal communications, or filing behavior show otherwise, the walk-away door slams shut.

This isn't unprecedented. Delaware has been tightening the standard for what counts as "commercially reasonable efforts" for years, increasingly rejecting buyer attempts to use regulatory conditions as a cover for buyer's remorse. But the Xactware Jira detail makes this one of the more vivid examples of a judge looking past the press release and into the company's own work records to figure out whether it was acting in good faith.

The implication for deal-makers is mechanical, not philosophical. If you're structuring an acquisition facing regulatory review and you know the FTC timeline might stretch past your outside date, you don't treat the regulatory effort as a formality. You push your teams to file, respond, engage, and document that engagement — because those records become the evidence in the lawsuit you're trying to avoid.

What the market sees now is a company being ordered to complete a large cash acquisition under court authority. Whether the deal is accretive or dilutive depends on what the integration looks like and how much Verisk actually values AccuLynx's data network within the restoration ecosystem — a question the company seems to have been reluctant to answer for itself. The structural judgment is simpler: the exit clause that was supposed to protect the buyer became a trap because the buyer's own internal records showed it hadn't earned the right to use it.