VERSATA SOFTWARE, LLC v. FORD MOTOR COMPANY
Authored by: Jeremy J. Gustrowsky
Ford Motor Company hired Versata (formerly Trilogy) to build software that would help it configure vehicles more efficiently. Versata delivered two programs: the Automotive Configuration Manager, or ACM, and the Materials Cost Analytics, or MCA. Ford licensed the software beginning in 2004 under a Master Subscription and Services Agreement. When that agreement was set to expire in 2014 and the parties could not agree on renewal terms, Ford rolled out its own configuration software, called PDO, which it had built while still licensing Versata’s products. Ford went to court first, seeking a declaration that it had not infringed Versata’s intellectual property or taken its trade secrets. Versata counterclaimed for trade secret misappropriation under the federal Defend Trade Secrets Act and the Michigan Uniform Trade Secrets Act, plus breach of contract.
Versata’s trade secret theory rested on three interlocking “combination” trade secrets within ACM: “Grid” (the user interface that shows how configuration rules interact), “Buildability” (the computing engine that generates workable part combinations), and “Workspaces” (which lets employees around the world collaborate in a shared environment). Versata also claimed the MCA, which converts configuration data into part costs, was misappropriated. Before trial, the district court threw out the testimony of Versata’s damages expert and ruled that trade secret damages had to be measured against the parties’ actual licensing history. That ruling knocked out two of the three replacement royalty models Versata later submitted, leaving only the one tied strictly to what Ford had historically paid. A jury then found that Ford breached the agreement and misappropriated the three ACM trade secrets (but not the MCA), awarding roughly $22.4 million on the trade secret claims and about $82.3 million on the contract claim. On post-trial motions, the district court left liability intact but cut the trade secret award to zero and the contract award to three dollars.
The Federal Circuit held that the district court made a legal error by shutting the door on unjust enrichment damages. Both statutes say plainly that recovery may include actual loss plus any unjust enrichment not already captured in the actual loss figure, with a reasonable royalty available as an alternative. The court noted that the Michigan statute reads almost identically to the Kentucky, Utah, and Missouri trade secret acts, and that the Sixth, Tenth, and Eleventh Circuits have all recognized unjust enrichment as an available remedy under that language. Quoting the Tenth Circuit’s reasoning, the panel observed that even if unjust enrichment leaves the owner better off than a license would have, the party that chose to take rather than license must bear the risk of that choice.
The panel also disagreed with the district court’s reading of Sixth Circuit precedent. Two older decisions had upheld royalty-based awards built on the parties’ dealings, but neither held that an owner is barred from seeking unjust enrichment in the first place. Ford argued the trial court had only excluded speculative unjust enrichment theories, not the category as a whole. The Federal Circuit was not convinced, pointing to the district court’s own language requiring that damages “be measured with reference to the parties’ licensing history” and its rejection of two royalty models simply because they looked beyond that history. The court vacated the ruling that zeroed out the trade secret award, sent the trade secret damages issue back for a new trial, and directed the district court to take another look at the two royalty models it had excluded.
On the contract claim, the Federal Circuit reversed outright and put the jury’s $82,260,000 award back in place. Michigan law requires damages proved with reasonable certainty, not mathematical precision. Versata had given the jury three annual figures drawn from the parties’ dealings: $17 million (Versata’s post-expiration license offer), $14.95 million (what Ford paid in the final year, including support and maintenance), and $10.95 million (the base license fee without support). Counsel told the jury to multiply its chosen figure by the 7.5 years at issue. The verdict worked out to roughly $10.97 million per year, sitting at the low end of a proof range running from about $82.1 million to $127 million. Versata’s testifying expert had broken down the components of the base fee, both versions of the contract were in evidence, and Ford’s own damages expert had conceded that the $10.95 million figure was a “much more reasonable starting point.” That was enough to give the jury a discernible path to a number.
Finally, the court rejected Ford’s cross-appeal on liability. Ford argued that Versata never showed Ford knew about each of the specific bundles of features making up the combination trade secrets, having instead handed over hundreds of documents containing them. The Federal Circuit found no such requirement in either statute. Both define misappropriation in terms of acquiring or using a trade secret through improper means or under circumstances creating a duty of secrecy, and neither demands proof that the defendant understood the exact combination. The Sixth Circuit had already refused to require a showing of acquisition of “each atom of a combination trade secret,” and the panel saw no reason to add one here. Testimony from Versata’s technical expert and a former employee described disclosure of the combinations through user guides, manuals, technical documents, emails, and on-site discussions with Ford’s engineers, which was enough to support the verdict.
For companies that license software or technology to a customer who later builds a competing product in-house, the practical takeaway is significant. A licensing history may be strong evidence of value, but it does not become a ceiling on recovery. Trade secret owners retain a statutory choice among actual loss, unjust enrichment, and a reasonable royalty, and a court cannot restrict them to the last option as a matter of course.