IP Due Diligence in Mergers: 5 Steps to Protect a Deal

Roughly a quarter of failed acquisitions trace back to problems the buyer could have caught before closing, and intellectual property is one of the most common blind spots. When the target’s value lives in patents, software, brands or trade secrets, a broken chain of title or an unlicensed open-source component can quietly erase the premium you paid. IP due diligence in mergers and acquisitions is the process that surfaces those risks while you can still price them, renegotiate or walk away. This playbook lays out the five buy-side steps that separate a clean deal from an expensive surprise.

Why IP Due Diligence in Mergers Decides Deal Value

In a technology or brand-driven deal, most of the purchase price buys intangible assets you can’t kick the tires on. This review is how the buyer confirms those assets are real, owned, unencumbered and enforceable before the money moves. Advisers routinely tie a large share of failed integrations to technical and IP oversights, and the fix is almost always cheaper before closing than after.

The findings do double duty. They tell you whether to proceed and at what price, and they anchor the representations, warranties and indemnities in the purchase agreement. A gap you miss in diligence becomes a warranty you can’t enforce.

Scope should track the deal. A software acquisition lives or dies on code ownership and open-source hygiene; a consumer-brand purchase turns on trademark registrations and clean chain of title; a life-sciences target is all about patent term, freedom to operate and licensed platform technology. Match the depth of review to where the value actually sits, and start early enough that what you find can still move the price.

Step 1: Verify Ownership and an Unbroken Chain of Title

Start with a question the seller often can’t fully answer: does the target actually own what it’s selling? Build an inventory of every material patent, trademark, copyright, domain and trade secret, then trace title from the original inventor or author to the company.

• Confirm employee and contractor IP assignments exist and are signed. Inventions created without a written assignment may still belong to the individual, not the company.
• Check that patent and trademark assignments were recorded with the USPTO. Under 35 U.S.C. § 261, an assignment can be void against a later bona fide purchaser if it isn’t recorded within three months.
• Verify that maintenance fees, renewals and registration statuses are current and nothing has lapsed.
• Reconcile foreign counterparts, since a U.S. registration says nothing about rights in the target’s key export markets.

Chain-of-title defects are the most common finding and the most fixable, provided you catch them before closing rather than after.

Step 2: Surface Encumbrances, Licenses and Open-Source Risk

An asset the target “owns” may already be promised to someone else. Read the fine print for anything that limits, burdens or terminates the rights you’re buying:

• Inbound and outbound licenses – check for exclusivity, field-of-use limits and change-of-control clauses that can terminate rights the moment the deal closes.
• Security interests and liens – a lender may hold a UCC-1 filing against the IP that has to be released.
• Open-source software – audit the code for copyleft licenses such as the GPL that can force disclosure of proprietary source. Unmanaged open source is one of the most common and most expensive findings in software deals.
• Standard-essential patents and FRAND commitments – these carry licensing obligations that travel with the patent to the buyer.

Each encumbrance is either a price adjustment or a condition the seller must clear. None should be a surprise after signing. Build a single schedule that lists every license, lien and obligation against each asset, and reconcile it against what the seller represented in the data room.

Step 3: Test Validity, Scope and Freedom to Operate

Ownership is worthless if the rights can’t be enforced or if using them infringes someone else. Pressure-test the portfolio on two fronts:

• Validity and status – look for pending litigation, PTAB inter partes reviews, trademark oppositions and prior-art threats that could knock out a key claim.
• Freedom to operate – confirm the target’s flagship product doesn’t infringe a third party’s patent, which is a separate question from whether the target owns its own IP.

A focused freedom-to-operate search on the deal’s core products often changes how a buyer values the whole transaction. Weak or narrow claims discovered here belong in the price, not in a post-closing dispute.

Scope matters as much as validity. A patent that reads on last year’s product but not the roadmap is worth far less than the seller thinks, and a trademark registered only in the home market leaves the brand exposed everywhere the target plans to grow. Read the claims and registrations against where the business is actually headed, not just where it has been. That forward look is what turns a checklist into real diligence.

Step 4: Diligence AI-Generated and Data Assets (New for 2026)

Two categories now demand their own review. First, AI-generated content: works created without meaningful human authorship generally can’t be registered for copyright, a line the courts reinforced in Thaler v. Perlmutter. If the target’s product catalog, code or brand assets were machine-generated, confirm what is actually protectable. Our breakdown of AI and copyright in 2026 covers where that line sits today.

Second, data. Verify the target has the rights and consents to use its training and customer data, that inbound licenses permit the intended use after the sale, and that its data practices comply with applicable privacy law. In a modern deal, the dataset can be the asset, and a broken consent chain can be as damaging as a broken title chain.

These questions are new enough that many sellers haven’t documented the answers, so expect gaps rather than clean files. Where the target can’t show a clear right to use an AI model, a training set or a third-party API, treat it the same way you would a missing patent assignment: a risk to price, escrow or condition, not something to wave through on a promise.

Step 5: Translate Findings Into Deal Terms

Diligence is only useful if it changes the paper. Map each material finding to a mechanism in the purchase agreement:

• Purchase-price adjustment – discount for lapsed rights, weak claims or licensing gaps.
• Escrow holdback – set aside part of the price against specific IP risks until they clear.
• Special indemnities – carve out known issues, such as a pending infringement claim or an open-source exposure, for dollar-one indemnification.
• Reps and warranties – require the seller to warrant ownership, non-infringement and disclosure of all licenses; consider representations-and-warranties insurance on larger deals.
• Conditions to closing – require cured assignments or recorded titles before the deal funds.

The worst outcome is a clean report that quietly ignored a known risk. If you found it, price it or paper it. And watch for the red flags that should slow any deal: missing employee assignments, key patents nearing expiration or with unpaid maintenance fees, core products built on copyleft open source, change-of-control clauses that kill critical inbound licenses at closing, and infringement litigation the seller downplays. Any one of these can justify a holdback; several together can justify walking away.

How PerspireIP Can Help

PerspireIP supports acquirers and their counsel with the technical backbone of IP due diligence in mergers and acquisitions: patent landscape and freedom-to-operate searches, portfolio and chain-of-title review, and invalidity analysis on the assets that matter most to the deal. Talk to our team before your next transaction closes.

Frequently Asked Questions