Patent Invalidity Search: What It Is and When You Need One

A patentability search asks "can I patent this?" A patent validity search, often called an invalidity search, asks a different question: "should this patent have been granted at all?" You run one when a patent already exists and something important hangs on whether it can survive a challenge. That usually means a lawsuit, a licence deal, or a demand letter sitting on someone's desk.

Getting this wrong is expensive. Miss the one reference that kills a claim, and you pay to license, settle, or lose a case you didn't need to. Find it, and a granted patent can come apart entirely. This piece covers what a validity search does, when you need one, what it costs, and how it differs from the patentability search you might already know.

When you actually need one

Most people never think about a patent's validity until a specific event forces the question. Four situations come up again and again:

You've received an infringement claim or a demand letter. Someone believes you're using their patented invention and wants you to pay, stop, or both. Before you negotiate, license, or lawyer up for trial, it's worth finding out whether the patent being waved at you would actually survive scrutiny. A shaky patent changes the whole conversation.

You're weighing an IPR petition. Inter partes review (IPR) is a process at the USPTO's Patent Trial and Appeal Board (PTAB) that lets you ask the Board to cancel a patent's claims using prior art — patents or printed publications. It's often faster and cheaper than fighting validity in federal court. But you typically get one shot at a petition covering a given patent and set of claims, so the search behind it has to be as complete as you can make it.

You're acquiring or licensing a patent. Buying a patent, or a whole portfolio, without checking whether the claims would hold up is like buying a car without checking if it starts. A validity search before the deal tells you what you're really paying for. If you're new to searching in general, it's worth first reading how a standard prior art search works, since a validity search builds on the same idea, just pushed much further.

You're auditing your own portfolio. Companies sometimes run validity checks on their own granted patents, quietly, to see how exposed they are if a competitor ever challenges them. Better to find the weak spot yourself than have a defendant's lawyer find it for you.

Here's a composite example of how this plays out. A small kitchenware manufacturer received a letter accusing one of its bestselling gadgets of infringing a patent held by a much larger rival. The instinct was to settle quickly and move on; legal fees alone looked painful. Instead, the company commissioned a validity search on the asserted patent. The search turned up a trade catalogue from years before the patent's priority date (the earliest date the invention can claim as its own), showing a near-identical mechanism already on sale. That single piece of evidence was enough to file an IPR petition. The case settled on far better terms once the patent owner saw the reference. Nothing about the manufacturer's product changed; what changed was the strength of the patent standing over it.

How this differs from a patentability search

A patentability search is a screen. You're checking a new idea against everything published before, looking for reasons an examiner might reject it. If you find a handful of close references, that's useful information, and you adjust your claims or your expectations accordingly. Missing one obscure reference rarely sinks the whole project, because the goal is realistic, not exhaustive.

A validity search is not a screen. It's closer to an evidence hunt with a court, or the PTAB, as the eventual audience. The logic is different too: you only need one reference that anticipates the patent's claims (shows every element of at least one claim) or, combined with another reference, makes the claimed invention obvious. Either can be enough to invalidate it. So in one sense the bar is lower, a single good document can do the whole job.

But there's a catch. You need to be reasonably sure you've found the best reference, not just a reference, before you build a strategy around it. At the PTAB especially, you rarely get a second try. Rules on estoppel mean that once you raise a ground in an IPR, or could reasonably have raised it, you generally can't bring it up again later in the same dispute. That's why an invalidity search has to be exhaustive rather than merely thorough: one missed reference at the search stage can mean one lost chance at the tribunal stage, with no do-over next month.

As one patent search specialist put it: "A patentability search is about giving reasonable comfort before you spend money on drafting. A validity search is about making sure that when you walk into a room with a judge or the PTAB, you're holding the best card in the deck, not the second-best one you happened to find first."

What searchers actually look for

Because the standard is higher, the hunting ground is wider than a typical patentability search. Specific things searchers chase include:

• Art the examiner never cited. Every granted patent lists the references the examiner considered. A validity search assumes those were already checked and looks elsewhere, different databases, classification codes, and search terms than the examiner likely used.
• Non-English documents. Patents and papers published in Japanese, German, Chinese, Korean, and other languages are prior art everywhere, but they're costly to search properly and often under-covered by a quick patentability check. A missed foreign-language reference is a gap the examiner probably fell into too.
• Anything published before the priority date. The priority date is the cut-off; only material public before it counts as evidence. Searchers work hard to pin down exact dates on catalogues, articles, and web pages, because a document dated a few weeks too late is worthless.
• Product and sales evidence, not just patent literature. Trade catalogues, old manuals, conference posters, archived websites, and proof that something was sold or offered for sale can all count. This kind of evidence rarely shows up in a patent database, which is exactly why it's valuable, the examiner almost certainly never saw it.

Typical cost and timeline

Validity searches cost more than patentability searches, and it's worth understanding why before you get a quote. A standard patentability search might run a few thousand pounds or dollars and take one to three weeks (see our breakdown of patent search costs), largely because the searcher is looking for the closest handful of references and stopping once the picture is clear.

Litigation-grade or IPR-grade validity work is a different job. Searchers dig through non-patent literature, chase foreign-language art, sometimes commission translations, and hunt for product or sales evidence that might sit in an old trade show archive rather than any database. That takes weeks, not days, and the analysis afterwards, mapping each reference against every element of the challenged claims, takes real attorney and searcher time too. A properly done validity search supporting litigation or an IPR petition commonly runs into five figures, sometimes well into them, once claim charts and expert input are added. A quicker validity check for internal due diligence can cost less, but it carries less weight if the patent is ever actually challenged.

Comparison: search types at a glance

Where to start

If you're facing a demand letter this week, don't wait for the perfect search before you talk to someone. A patent attorney can tell you whether a full validity search, an IPR strategy, or a quicker settlement conversation fits your situation — our notes on attorneys versus DIY searching cover that trade-off. A preliminary search tool can give you a fast read on how crowded the art looks before you commission the deeper, exhaustive work a real validity search needs; it's useful for triage, not a substitute for it. Compare plans if you want to run a few checks yourself first.

Key takeaways

• A validity search (invalidity search) looks for prior art that could knock out an already-granted patent, usually because someone is enforcing it or about to.
• It differs from a patentability search in thoroughness, not subject matter: you need to find the single best reference, and be confident it's the best one available.
• Common triggers: an infringement claim or demand letter, considering an IPR petition, buying or licensing a patent, or auditing your own portfolio.
• Searchers dig into art the examiner never saw, foreign-language documents, and product or sales evidence, not just patent databases.
• Litigation-grade validity searches often run into five figures, well above a standard patentability search, because the stakes and depth of digging are much higher.

None of this is legal advice, and this article isn't a substitute for it. Whether a specific patent is vulnerable, whether an IPR petition suits your situation, and how to weigh a settlement offer are questions for a qualified patent attorney who knows your facts. Use a validity search to inform that conversation, not replace it.

Sources and further reading

• USPTO Patent Trial and Appeal Board (PTAB)
• USPTO Patent Public Search
• MPEP (Manual of Patent Examining Procedure)