No Swerving from Fraud: Delaware Tests the Post-Fortis Landscape

The Delaware Supreme Court in January 2026 held in Johnson & Johnson v. Fortis Advisors LLC¹ that a one-sided anti-reliance clause from the acquiror (agreeing not to rely upon extracontractual statements made by the seller or its representatives) will not shield an acquiror from a seller's post-closing fraud claim against the acquiror regarding extracontractual statements allegedly made by the acquiror. In the months since, the Delaware courts have been actively defining the contours of that holding – narrowing the implied covenant of good faith and fair dealing in earnout disputes while continuing to treat misrepresentations of present fact as actionable fraud against an acquiror. The Delaware Court of Chancery's July 31, 2026, decision in In re SwervePay Acquisition, LLC² (the Decision) is the most consequential data point yet in that developing body of law: a fully litigated, $120 million post-trial fraud verdict built almost entirely on the acquiror's own internal communications, showing precisely what Fortis-era exposure looks like once a case reaches the fact finder. The Decision carries lessons not only for litigators handling post-closing disputes, but for the corporate lawyers drafting the earnout provisions and representations that will be tested if a transaction later goes wrong.

The Transaction

In February 2020, a group of private equity sponsors (the Buyers) that already owned a debt-collection software platform (the Platform) acquired a small payment facilitator (the Target), intending to integrate the Target's technology into the Platform. In exchange, the Target's founders and executives (the Sellers) received approximately $9.9 million in cash and an additional $10 million in rollover equity at closing, together with contingent earnout consideration of up to $53.75 million tied to the Target's ability to convert the Platform's existing payments volume onto its own systems during the 2021 earnout period.

The value of that contingent consideration turned on a single, contested metric: the volume of payments actually flowing through the Platform's systems. The Buyers shared with the Sellers $34 billion of estimated payment volume, which was sourced from an investment bank's industry-benchmark estimate rather than the Platform's actual transaction data. The Buyers' payments analyst subsequently determined that the actual payment volume was only approximately $5.2 billion – roughly one-sixth of the amount the Buyers continued to represent to the Sellers as the actual volume throughout the negotiation.

As the court found, contemporaneous emails show the deal team debating internally, in real time, whether there existed "any intellectually honest way" to present the $34 billion payment volume to the Sellers before ultimately transmitting it without qualification hours later. One buyer-side executive's email the following morning candidly conceded that what had been sent was "meaningless" – "a red herring." The Court of Chancery found that exchange, among a substantial body of similar correspondence, dispositive of scienter necessary for a fraud claim.

The Holding: Present-Fact Fraud Survives an Absent Anti-Reliance Clause

In contesting the fraud claim, the Buyers advanced two principal defenses to reliance. First, they argued that the Sellers could not have justifiably relied on an unverified estimate. Second, they contended that the absence of any contractual representation regarding payments volume should itself weigh against a finding of reasonable reliance – reasoning that a sophisticated seller who wanted assurance on the point should have bargained for it. The court rejected both arguments. Under settled Delaware law, a defendant seeking to foreclose a fraud claim through contract must secure specific, unambiguous anti-reliance language. Absent such language, silence favors the party alleging fraud, not the party accused of it. Because the Purchase Agreement lacked an anti-reliance clause from the Sellers regarding extracontractual statements made by the Buyers and their representatives, and because the Sellers' diligence requests sought data that only the Buyers possessed, the court found the Sellers' reliance both actual and reasonable – expressly distinguishing precedent in which a plaintiff's own inadequate diligence, rather than an asymmetry of information, had defeated a reliance finding.

This is the same doctrinal architecture the Delaware Supreme Court endorsed in Fortis. The doctrinal line that determines which fraud claims are viable in the first instance is the distinction, introduced above, between present fact and future promise: A misrepresentation of present fact is actionable even where the agreement contains a standard integration clause – a provision stating that the written contract is the parties' entire agreement – because, as already discussed, only explicit anti-reliance language, not run-of-the-mill integration-clause boilerplate, can foreclose a fraud claim. A promise regarding future conduct, by contrast, generally is not actionable as fraud at all. The Court of Chancery drew precisely the present-fact line just months earlier in another case, where a buyer's pre-signing statement that it "already had budget approval" to expand a sales team survived a motion to dismiss as an actionable statement of present fact, while a companion representation regarding the buyer's future product-development plans was dismissed as a non-actionable promise of future conduct. This case fits squarely within the present-fact category Meyers describes: The Sellers sought data describing what payments volume existed on the Platform at the time of negotiation, not a promise about how the Buyers would operate the business going forward. Read together, previous precedence and the Decision illustrate that the present-fact/future-promise distinction, rather than the mere existence of integration-clause boilerplate, has become the operative battleground in post-Fortis earnout fraud litigation – and that Delaware courts are prepared to parse a single set of pre-signing representations into actionable and non-actionable components with real precision.

The Damages Wrinkle: The Court Built Its Own Model

Perhaps the most instructive aspect of the opinion, however, concerns not liability, but damages. Rather than choosing between the parties' competing experts – the Sellers' expert valued the equity-based earnout at between $103.1 million and $218.6 million, while the Buyers' expert argued for a materially lower amount – the court rebuilt the underlying valuation model itself, drawing several key inputs – including the growth rate and the Platform's stand-alone earnings before interest, taxes, depreciation and amortization (EBITDA) – from the Buyers' own internal documents rather than the Sellers' more optimistic projections. The resulting award – $75,692,297 – fell below the floor of the Sellers' own expert's range.

This methodology is not unique to the Decision; it reflects an emerging pattern in how the Court of Chancery approaches earnout damages more broadly. In another case decided little more than a year earlier, the court likewise declined to adopt either party's number, instead building its own expected-value model – traced directly to the Court of Chancery's trial-level decision in Fortis – and rejecting the sellers' request for full milestone payment as an impermissible "windfall," resulting in an award of $180,944,915. Delaware's rule that uncertainty in damages is resolved against the wrongdoer does not necessarily translate into deference to the plaintiff's expert; courts are exercising independent, granular control over each input, including the defendant's own documents, rather than picking a number from either side's presentation.

Drafting Takeaways

The Decision also offers concrete lessons for the corporate lawyers who draft the agreements litigators later have to litigate. Fortis and the Decision both turn on the same drafting gap: a missing, or one-sided, disclaimer of reliance on statements outside the four corners of the agreement. A mutual, bilateral anti-reliance clause – running against both parties, not just one – remains the most direct way to foreclose this exposure. Illustrative language:

No Other Representations. Each party acknowledges that it has not relied on, and will not assert any claim based on, any representation, warranty, statement, projection or other information of any kind, whether written or oral, made or provided by the other party or its Representatives, other than the representations and warranties of such party expressly set forth in this Agreement. Neither party nor any of its Representatives shall have any liability to the other party arising from the other party's use of, or reliance on, any such information not expressly set forth herein.

Sample language only. Must be reviewed and tailored to the specific transaction; not a substitute for legal advice.

Second, the Decision is also a reminder to check whether a "no interest" provision reaches only contract claims or extends to tort theories as well. The underlying agreement here specified that no interest would accrue on earnout payments, but the court held that language addressed only a breach-of-contract claim, not a claim sounding in fraud – and because prejudgment interest is available as a matter of right under Delaware law absent language to the contrary, the court awarded it on the fraud recovery over the Buyers' objection. Deal counsel wishing to close that gap should draft accordingly:

No Interest. No interest shall accrue or be payable on any Earnout Payment, and no party shall be entitled to prejudgment interest with respect to any dispute, claim or cause of action relating to any Earnout Payment, whether such dispute, claim or cause of action sounds in contract, tort (including fraud or fraudulent inducement), or any other theory of liability, at law or in equity.

Sample language only. Must be reviewed and tailored to the specific transaction; not a substitute for legal advice.

Third, notwithstanding the fraud verdict, the court denied the Sellers' request for attorneys' fees. Delaware follows the American Rule, under which each side bears its own fees absent a contractual fee-shifting provision or a finding that the losing party litigated in bad faith. Neither applied here: The Sellers had not sued under the one agreement that contained a fee-shifting clause, and the court found that the Buyers' defense, though unsuccessful, was not vexatious. The result often surprises clients more than lawyers – a nine-figure fraud verdict does not, on its own, shift fees in Delaware.

The Broader Landscape

Viewed alongside recent precedence decided only three weeks earlier, the Decision rounds out a fuller picture of the post-Fortis landscape. Recently, for example, the Delaware Superior Court's Complex Commercial Litigation Division applied Fortis' foreseeability rule to dismiss an implied covenant claim seeking to rebalance an earnout after an unanticipated but foreseeable contingency. Read together with the Decision and Meyers, the current state of Delaware earnout jurisprudence is coming into focus: The implied covenant is contracting as a tool for sellers to fill gaps the parties simply failed to negotiate, while extracontractual fraud claims – properly pled around present-fact misrepresentations and unprotected by anti-reliance language – remain a live and increasingly well-litigated avenue of recovery. Transaction counsel on both sides of the table would be well served to take note.

Holland & Knight routinely represents purchasers and sellers in post-closing mergers and acquisitions disputes and claims relating to earnouts, purchase price adjustments, and breaches of representations and warranties and indemnity, including litigation and trial of these issues in the courts of Delaware. Our deep bench of practitioners continuously monitors developments in the law affecting transaction participants.

Notes

¹ Johnson & Johnson v. Fortis Advisors LLC, 352 A.3d 229 (Del. Jan. 12, 2026).

² In re SwervePay Acquisition, LLC, Consol. C.A. No. 2021-0447-KSJM (Del. Ch. July 31, 2026).