Two Delaware Decisions Every Seller May Want to Read Before Signing

In September, the Court of Chancery ordered a buyer to provide biweekly pipeline and pricing data so a seller could monitor a $50 million earnout. In July, the Delaware Supreme Court revived a fraud claim arising from a $100 million acquisition that a trial court had rejected on the ground that the buyer had been willfully blind. Together, they may say something useful about what a purchase agreement is worth after closing.

Much of what happens after a deal closes is governed by a document the seller may have stopped reading closely somewhere around the fourth turn of the draft. Two Delaware decisions this year offer a reminder of what that document may, and may not, protect.

The first is about an earnout and what a seller may be able to demand when the buyer controls the data. The second is about what can happen when a seller conceals something and the buyer fails to find it. One favored the seller, one favored the buyer, and both seem worth ten minutes of a founder’s time before the next letter of intent. What follows is a high-level summary of two specific cases; how either would apply to a different agreement depends on its terms and facts.

Winton v. North Highland: the buyer may have to show its work

In December 2024, a technology consulting firm called The Bridge was acquired by North Highland, the U.S. operating entity of a global change-management consulting firm. According to the opinion, the letter of intent valued the company at $180 million: $90 million in cash, $40 million in rolled equity, and a $50 million earnout targeted over two years. The earnout keyed off “Qualifying Projects,” defined as new North Highland projects that “primarily (based on revenue)” involved technology or data analytics services. The seller representative was to flag potential projects before a proposal went out, and the parties were to “mutually agree” on classification.

That process appears to have broken down quickly. The court found that within two months of closing North Highland took the position that certain revenue would not count, that by late March it had circulated a “Ways of Working” document excluding certain clients and regions from earnout credit, and that it “withheld pricing tools” and “provided curated pipeline reports that omitted projects it unilaterally deemed ineligible.” The opinion recounts that on May 6, 2025, the buyer terminated the seller’s CEO, who was also the seller representative, purportedly for cause, without telling him the reason. He sued in January 2026.

Vice Chancellor Will bifurcated the case and held a two-day trial in May on the forward-looking questions. The September 18 post-trial opinion answered them largely in the seller’s favor.

On information: the court held that the implied covenant of good faith and fair dealing “prohibits North Highland from unreasonably restricting the data necessary to satisfy that notification condition,” and that the buyer “cannot demand strict compliance with this condition precedent while wielding its exclusive control over its systems to arbitrarily withhold the sole means of effectuating it.” The court did not give the seller live access to the buyer’s systems, and it described the implied covenant as a “limited gap-filler” that “may not be used to rewrite or contradict express terms.” But it ordered specific performance: biweekly reports of projects moving through the buyer’s sales pipeline, including the Salesforce fields needed to identify each project and the pricing-tool output showing the revenue split.

On the definition: the court held the Qualifying Project test was “complete, unambiguous,” and purely revenue-based, and that the buyer’s carve-outs “are foreclosed by the plain, unambiguous text.” The buyer could not, in the court’s view, use a defined term to add a cross-sell requirement the contract did not contain.

On the “mutual agreement” step: the court called it “a verification exercise,” not a veto. The buyer keeps its discretion over which projects to pursue, staff, or continue, but “once North Highland initiates a project that satisfies the Qualifying Project criterion,” the classification process is triggered and the buyer “cannot withhold its agreement based on extra-contractual criteria.”

The case is not over. The court denied without prejudice a request for separate business-unit P&L statements, the non-expedited claims remain pending, and the decision could be appealed. As of this writing we have not seen firm commentary on it. But the practical observation does not seem to depend on how the rest of the case comes out: the seller here had to win a trial to obtain information a single paragraph in the earnout addendum might have provided. The express information right the court discussed ran to working papers after an earnout statement was issued. Nothing the court pointed to addressed contemporaneous pipeline visibility, which may be the thing that actually lets a seller monitor the metric.

Paragon Metals v. Smith: “you should have caught it” may not be a defense to fraud unless the buyer deliberately looked away

The second decision runs the other way. In 2019, Stellex-affiliated buyers paid $100 million for Paragon Metals, an automotive parts manufacturer. Before closing, two of Paragon’s top three customers had told the seller they would materially cut orders. According to the trial court’s findings, as recounted in the Delaware Supreme Court’s opinion, the seller deleted a customer letter from an email attachment sent to the buyer, stripped volume terms from an amended contract, routed a rebate through a foreign affiliate, and never updated the projections. The buyer spent roughly a million dollars on diligence, saw a strike-through reference to a customer decrease, accepted the seller’s explanation, and did not ask for the letter.

The trial court found the representations false and the seller’s intent proven, then ruled for the seller anyway: the buyer had been “willfully blind” and its reliance was not justifiable. On July 1, 2026, the Supreme Court reversed that part. Willful blindness, the court said, requires that the buyer “subjectively believe that there is a high probability that a fact exists” and “take deliberate actions to avoid learning of that fact.” Missing red flags, on this reading, is not enough. In the court’s words, “Stellex’s trust in Smith, while perhaps naïve, did not amount to deliberate action to avoid discovering the truth.” And, quoting the Court of Chancery’s 2022 Arwood decision, a “failure to discover the fraud through due diligence does not excuse it.” The case went back for damages.

The court also read the purchase agreement’s “independent investigation” clause, the standard language in which a buyer acknowledges it did its own diligence, as a waiver of the buyer’s right to rely on extra-contractual statements, not as an affirmative promise to diligence well. As Reed Smith put it in August, “anti-reliance clauses protect sellers from extra-contractual claims,” but “they do not impose affirmative investigation obligations on the buyer to conduct objectively reasonable due diligence.” Sidley’s September summary was that the decision “confirms that a seller cannot avoid liability for knowingly false representations by arguing that better diligence by the buyer would have revealed the truth.”

A note on what this case appears not to be. It is easy to mistake for a “sandbagging” decision. It does not seem to be one, and the opinion does not use the word. Sandbagging is generally described as the contract question of whether a buyer who knew a representation was false at signing can still sue for breach; in the 2022 Arwood decision, one Vice Chancellor wrote that “in my view, Delaware is, or should be, a pro-sandbagging jurisdiction,” and the Supreme Court does not appear to have squarely resolved the question. Paragon looks like the fraud-side complement: a buyer’s failure to discover a falsity may not defeat a fraud claim unless the buyer actually knew or deliberately avoided knowing. Read together, the two decisions may suggest the “you should have caught it” argument is getting narrower from both directions, though the sandbagging half remains unsettled and actual knowledge and true willful blindness remain available defenses.

Why this may matter now

Market data may explain why both decisions resonate. According to the ABA’s 2025 Private Target Deal Points Study, roughly seven in ten purchase agreements in the sample were silent on sandbagging, and only about 4 percent contained an anti-sandbagging clause. SRS Acquiom’s 2026 study reported earnouts in about 24 percent of the 2025 deals it tracked, with the median earnout potential rising to about 34 percent of the closing payment. In other words, many sellers appear to be signing agreements that say nothing about buyer knowledge, and a meaningful share are accepting contingent consideration that depends on numbers the buyer will control.

Questions that may be worth asking before the LOI

• If there is an earnout, what information would you actually receive, how often, and before or after the decisions that drive the metric? (Winton’s seller had working papers after the fact; what he appears to have needed was the pipeline before proposals went out.)
• Is the earnout metric defined in a way that leaves little room for “operational” carve-outs, and does the agreement say what happens when the parties disagree on classification?
• On the seller side, do the disclosure schedules say plainly what the seller knows about customer changes, litigation, and projections? Paragon suggests that hints and strike-throughs may not protect a seller who knew the answer.
• Does the agreement address sandbagging at all, and if not, is silence the outcome you want?
• Who is the seller representative after closing, and what happens to the earnout process if that person is terminated?

This article summarizes two court decisions at a high level and omits procedural details and open questions; Winton remains pending on non-expedited issues and could be appealed. It is a general description, not a complete statement of Delaware law, and is current as of October 15, 2026. Market statistics are as reported by the cited studies.

How CGL can help

CGL represents sellers in negotiating earnouts, representations, and post-closing protections, and in our experience we tend to be most useful before the letter of intent locks the structure in.

If you are reviewing an offer with an earnout in it, reply with the word “earnout” and we will send you a short list of the earnout terms that, in our experience, often shape whether the contingent piece pays. Just the one word is enough; we will follow up.

CGL LLP · Los Angeles, California · Attorney Advertising

This newsletter is provided by CGL LLP for general informational purposes only and is not legal or tax advice; reading it or replying to it does not create an attorney-client relationship. The information is current as of October 15, 2026 and may change. Please consult counsel about your specific situation.

• Corporate Insights
• Startup Advice
• Weekly Briefs