Picture a Monday morning. You open your inbox and find a letter from a competitor's lawyers. It says your best-selling product uses an invention they hold a patent on, and they are taking you to court.
Your stomach drops. You start counting what this could cost in legal fees, in royalties, and in lost sales. Your team wants to know what happens next, and you don't have a quick answer.
Here is the part many people don't know. A granted patent is not untouchable. If something already existed in the world before the patent's earliest filing date that shows the same invention, the patent's claims can be challenged. That something is called prior art. The work of finding it, organizing it and presenting it is what this guide is about.
In this article we follow a hypothetical customer through the whole journey. We call them Company B, and the competitor suing them Company A. Along the way we cover what the search is, who needs one, how it works step by step, what it costs, and how to choose who does the work. Nothing here is a real customer's story. It is a plain-language walk-through of a process that real companies go through.
If you read only one idea from this guide, make it this one: the lawsuit is not the end of the conversation. It is the start of a research project, and the quality of that research can change the outcome.
First, What Exactly Is This Kind of Search?
Let's start with the basics, because the jargon around patents can make simple ideas sound scary.
A patent invalidity search is a prior art search carried out on a patent that has already been granted. The patent being examined is usually called the subject patent. The goal is to find evidence that the patent's claims should never have been granted, because the invention wasn't new, was obvious, or lacked an inventive step.
You will hear this called by a few names. Some people say invalidation search, and others say opposition search. They all mean the same thing: digging for earlier material that undermines the claims of a granted patent. If you hear someone talk about patent invalidation, this is the kind of work they are referring to.
The searchers look for two kinds of material. The first is patent literature, such as earlier patents and patent applications. The second is non-patent literature, which covers everything else that was made public. Both are valuable because they can be the very things the patent examiner overlooked when the patent was originally reviewed.
Why would an examiner overlook something? The source material gives practical reasons: time constraints, a backlog of applications, and the simple fact that some material wasn't available to them. Examiners are people doing a big job under pressure. A second, focused look from a team dedicated to one patent can turn up what the first pass missed.
There is one rule that matters more than any other. To count, the prior art has to have been published or made available to the public before the subject patent's earliest filing date. Something that appears after that date, however relevant, can't be used against the patent. Keep that date in mind, because it comes back again and again in the process.
Two Sides of the Same Coin: Invalidation and Validation Searches
Our imaginary Company B is in the hot seat, so it wants to knock the patent down. But the same kind of search also helps people who want to confirm a patent is strong. The steps are very similar. The only real difference is who is asking and why.
An invalidation search is done for someone who has been sued, the defendant. Their plan is to fight the lawsuit by going after the patent it rests on. If the patent's claims fall, the case has nothing to stand on.
A validity search, sometimes called a validation search, is done by the patent's own assignee. They want to know whether the patent portfolio is strong, and how likely it is that the patent would survive if somebody challenged it in a lawsuit in the future. Think of it as a stress test carried out before anyone else gets the chance to run a real one.
In practice, these searches show up in a handful of recurring situations. Let's walk through them, because they show how many different people end up needing the same kind of research.
A defendant wants a defense. A patent owner, the plaintiff, sues a competitor for using a protected invention without permission. The defendant turns to the search to show the patent's claims are invalid. If the grant is declared invalid, there is no infringement case. This is why the search works as a defense against infringement claims.
A patent holder wants to be sure before suing. Before filing against an alleged infringer, the owner may order a validity search. If the search finds no satisfactory prior art, the claims are assumed to be novel, non-obvious and inventive, which gives the owner more confidence that going to court is sound.
A company is thinking about licensing. Someone who would need a license to make or sell products covered by the claimed invention may order a validity search first. Licensing takes serious money and time, and the search helps the potential licensee judge the strength of the assignee's patents before committing.
A company is considering buying another company. When Company 1 wants to acquire Company 2 mainly for its patents, it asks for a valuation of the patent portfolio and a validity search on it. A patent with valid claims is worth more, so the search feeds directly into the price.
Portfolios can also be ranked on factors like the number of cited references, the size of the invention families, the number of independent and dependent claims, litigation history, reassignment and so on. The higher a patent ranks, the stronger it is considered to be, and the more it is worth.
For Company B, though, none of that is the immediate concern. B is the defendant. B needs the invalidation side of the coin.
Company B's Choices: Settle, Fight, or Fight Smarter
Let's go back to our scenario. Company A believes that Company B's product uses A's patented invention, and that B is earning money from it. So A has filed a lawsuit.
Company B now faces two obvious options. The first is to settle by paying royalties to Company A. The second is to fight and argue that B's product is not like the patented invention.
There is a third angle, and it is the one that makes this kind of research so powerful. The plaintiff's whole case can fall apart if the patent itself is invalid, if it should not have been granted in the first place. If prior art exists from before the patent's filing date, the patent can be nullified, disabled or killed, to use the terms in the source material.
Think about what that means for Company B. If B can show the patent is invalid, B does not owe royalties, even if B's product happens to look like the patented one. The argument shifts from whether B's product matches the patent to whether the patent deserved to exist at all.
This is why the defendant hires experts to research the patent in question. They are looking for prior art that discloses what the claims describe. If they find it, it may protect B from the financial damage a lawsuit could cause. The source material notes that winning this fight could save a company millions of dollars.
That does not mean it is easy. It is hard to invalidate a patent that has already been the subject of multiple legal proceedings. But even on a tough case, the search is often the single most important piece of preparation a defendant can do. Without it, the defendant is left arguing only about how its product differs from the patent. With it, there is a second line of defense.
There are also routes beyond the courtroom. Inter Partes Review (IPR) and Post Grant Review (PGR) petitions can be filed against a granted patent before the Patent Trial and Appeal Board (PTAB). These challenges are brought on various grounds, including the submission of prior art. A search is how you find the prior art to submit, and the person filing can run it on the very patent they plan to challenge.
Companies file these petitions for different reasons. A company working in a technology area may see a direct competitor granted a patent in a similar domain, and may want to gain a stronger hold in the field by challenging it. A company that wants to launch a product that would infringe the claims of a patent might file one too. In every one of those cases, the work begins with a search.
The Rules of the Game: Dates, Laws and What Counts as Prior Art
Before Company B's team runs a single search, they need to understand what counts as usable evidence. This is where many people get tripped up, so it is worth going slowly.
The all-important cut-off date
Every search has a cut-off date. In most cases it is the subject patent's earliest filing date, also called the earliest priority date. Prior art has to have been published, used, known or sold before this date to be valid in a court of law.
A simple way to picture it: draw a line on a calendar at the earliest filing date. Anything that was public before the line is fair game. Anything after the line is off limits, however perfectly it matches the claims.
There is one exception worth knowing about. If the subject patent is a Continuation-in-Part, or CIP, of another application, the analyst has to look carefully. If some part of the target claim matches new description text added in the CIP, then that part of the claim gets the CIP filing date as its cut-off. The rest of the claim keeps the earliest priority date. And if the whole target claim falls within the newly added description, the CIP filing date becomes the cut-off for the entire claim.
So in a CIP case a single claim can end up with two different cut-off dates, one for each part. That is the kind of detail a careful analyst catches and a rushed one might not.
Which law applies
For a United States patent, the analyst also has to know which version of the law applies. If the search cut-off date is before March 16, 2013, the Pre-AIA rules apply. If it is on or after March 16, 2013, the Post-AIA rules apply.
Different jurisdictions have their own sets of rules too, such as US laws and EP laws. These decide whether a given piece of prior art can be used to invalidate a patent. So the people doing the search need to know those laws well. For a US subject patent, USC 102 has to be considered when analyzing prior art.
What can be prior art
The range is wider than most people expect. Patents and patent applications count. So do non-patent citations, which can include:
Research papers and theses
Standards and white papers
Conversations and e-mails
Newsletters and articles
Products, including proof of sale
Videos, blog posts and any other internet publications
A product on the market before the cut-off date can be prior art. So can a blog post. What matters is that it was available to the public in time and that it shows what the claims describe. This is good news for a defendant, because the evidence you need may not sit in a patent database at all. It may be in an old product listing, a conference paper or a forgotten web page.
The grace period twist
Inventors and applicants in the United States get a one-year grace period after publicly disclosing their own invention to file a patent application. So if the inventor's own document was published within a year before the filing date, it is not considered prior art. But if that document, or another one by the same inventor, exists more than a year before the filing or priority date, it can count.
There is a further wrinkle. If a document comes from someone with at least one inventor who is different from the subject patent's inventors, the grace period does not apply. That document does not need to be a year older. As long as it was filed before the search cut-off date, it can serve as prior art.
One more note from the source: to be used together to invalidate a patent, the citations must be published prior to the search cut-off date. Every piece in the combination has to clear the line.
What the search is testing
This kind of search asks whether an issued patent is valid under the criteria of novelty, non-obviousness and usefulness. The search looks for prior art that discloses a claim limitation, rather than the entire subject invention. That is a subtle but important point. The team is hunting for evidence about the specific wording of the claims.
The grounds for a search include prior public data or use, publication, or sale of the patented technology, process, product or service before the priority date. The priority date is the earliest filing date of a patent at a patent office such as the USPTO, EPO, WIPO or IPO India. The grounds also include objections or rejections made by the examiner during prosecution. In the United States, for example, those are 35 U.S. Code 101, 102 and 103 rejections.
How a patent actually falls
Here is a point that surprises a lot of first-time clients. To invalidate a claim, the analyst needs either a single prior art reference that covers all the claim's features, or a combination of at least two references that together cover them.
And invalidating one claim is not the same as invalidating the patent. If a single independent claim falls, only that claim is invalidated. All the other claims remain in force. To invalidate the entire patent, every claim would have to be invalidated. That is why the choice of which claims to target matters so much.
For Company B, this means the team and the lawyers should agree early on which claims Company A is actually asserting. There is no point in spending the budget on claims that are not part of the lawsuit.
Day One for Company B: Building the Team and Understanding the Patent
Company B has now hired a firm to run the search. What happens first? This is the part of the process that clients rarely see, and it explains a lot about why a good search takes the time it does.
Assigning the right people
When the firm receives the request, the first step is to look into the domain of the patent and all of the client's instructions. Then a team of experts is assigned. The team has a hierarchy that runs from a manager down to research analysts. Who is on it depends on the domain, and on practical factors like the committed submission dates, the budget and the allotted hours or days.
The search also starts with picking a team that is familiar with the patented invention and setting the search cut-off date. A team that already knows the field will understand the technical language of the claims faster and spot relevant documents more reliably.
Reading the patent properly
The analyst's first job is to understand the patent and its claims. Which claims get searched depends on the client's needs. It could be:
The most general independent claim
All independent claims
All asserted claims, meaning the ones put forward in a lawsuit that is already underway
Only the dependent claims
For Company B, the asserted claims are the natural focus. Those can be found in the complaint letter filed in the lawsuit, which is where the infringement contentions are set out. Complaint letters are available from several sources, including RPX and the USPTO.
Understanding the patent follows a set order. The analyst reads the title, abstract, summary and claims. Then they go to the description to get to grips with the claim's difficult or narrow parts. Scope matters here. These searches are always conducted on the claims of the subject patent, with little emphasis on the subject matter and embodiments in the description. But the claims are interpreted in the light of that description, so the description still helps explain what the claim words mean.
The file history
The next step is one of the most useful. The file history, also called the file wrapper or prosecution history, holds all the drafts, the applicant's arguments against the examiner's rejections, notices of allowance and more. It is how the analyst understands what made the claim novel in the eyes of the patent office.
For US and WO patents and applications, the file and prosecution history can be obtained from USPTO Public Pair or Espacenet.
The file history also reveals the arguments the applicant used to defend the patent. If the applicant fought hard to show that one feature set the invention apart, that feature is a clue to where the search should dig. It shows the examiner's own search strategies too, which can give the analyst ideas for search strings and some valuable keywords.
Getting everyone on the same page
After the analyst has built their understanding, it is discussed with the manager or team head and with the client. This is where misunderstandings get caught early, before any budget is spent in the wrong direction.
Clients can share a lot at this stage. Company B may have already run its own preliminary search and found a few claim elements. If it shares them, the searchers can leave those known citations out and concentrate on the parts of the claim not yet found. B may also know of companies, assignees or players who are very active in the field. Telling the searchers lets them look closely at exactly those names.
The Hunt Begins: How the Search Actually Works
With the groundwork done, the real searching starts. This is the part that sounds the most technical, so let's keep it practical.
Step one: the exclusion list
The analyst first creates an exclusion list. This is a list of citations, both patents and non-patents, that are already cited on the face of the subject patent, along with their family members. Generally these are left out of the search because the examiner has already seen them. A client's needs can change that, though.
This list includes citations the examiner raised during examination, which are also discussed in the file history. Studying them gives valuable insight into the technology and into which kinds of citations are not worth chasing again. In effect, the exclusion list tells the team where the examiner has already looked, so they can spend their effort elsewhere.
Step two: quick first-pass searches
Next come the first-pass strategies. The team can start with free automated searching on Google or Google Patents. They can also use AI-based tools. The source material describes one where you enter the patent number to be invalidated and it automatically generates a bucket of patents.
The first-pass options include citation analysis, similar searches, and family analysis in other jurisdictions along with prosecution history analysis.
Citation analysis looks at two sets of patents. Set A is the forward citations of the backward citations of the subject patent. Set B is the backward citations of the forward citations of the subject patent. Those two sets can be combined with OR or AND depending on how many results there are, how much time the search has and similar factors. The labels A and B are just descriptive and can be swapped.
The bucket is then narrowed by applying the search cut-off date, along with a few classes or search keywords. Citation analysis can be done on any database.
Family analysis in other jurisdictions sounds absurd when you read how it is built. It is a data set that includes the backward citations of the subject patent and their family members, plus the backward citations of the backward citations of the subject patent's family members. But it makes sense for a simple reason. A family member of the subject patent may have one or more unique backward citations that the subject patent does not. Those citations are not cited on the face of the subject patent, so they could be used to invalidate it. In this search, the patents on the exclusion list are removed and the rest are analyzed for relevant prior art.
A similar search is simply a collection of patents in the same domain of technology as the subject patent. They may well be useful.
These first-pass methods do more than produce documents. They give the analyst deeper insight into the field, along with keywords, synonyms and classes that can steer the rest of the search. They can also show which assignees matter most and which parts of the claim have not yet been found anywhere.
Step three: keyword and class searches
Then the analyst moves on to strings. A collection of search strings is built and applied to many databases. The strings are created from several ingredients:
Technical and non-technical keywords, and their synonyms
International classifications such as IPC, CPC, ECLA and FI/F-terms
The most active players in the domain
Other strategies
The strings cover the claim scope from narrow to broad, using all possible methods, keywords, strings and classes. Starting narrow finds the closest matches quickly, and widening step by step makes sure nothing is missed on the edges.
Databases come in both paid and free varieties. The source mentions Orbit, Derwent, Patsnap, Google Patents, Lens.org, Espacenet and J-PlatPat among others, as well as Questel Orbit, Lexis Nexis (Total Patents), and Google Scholar and IEEE Xplore for non-patent work.
There is also a newer approach. Automated searching combines intelligent patent parameters such as classification, citations and assignees with Natural Language Processing models. The result is a quick search that uncovers results on the basis of contextual and expert parameters.
Step four: non-patent and standards searches
A patent search alone is rarely enough. Non-patent searches are run on databases like Google, IEEE Xplore, Science Direct, and Springer. As with patents, non-patent prior art has to be published or made available to the public before the cut-off date.
A standards-based search can also be done, on standards databases such as IETF, ETSI and 3GPP, depending on the scope of the invention. This matters especially in some industries. The source notes that in telecom and wireless patents, basic searching plays a significant role, and that these searches span databases from all over the world, not just the United States.
Step five: work the bucket
After these searches, the analyst has a bucket of patents and non-patent citations. They don't just read it once. They can put it to work in several ways:
Pull out the top assignees and inventors, and search for them specifically
Analyze the top patent and non-patent references cited on the face
Extract classes from these patents and use them in new search strategies
Run the automated searching strategies again on the most closely related references
These searching procedures keep changing depending on client requirements, project budget, the domain of the subject patent and so on. What was learned from each pass shapes the next one, so the search tightens as it goes.
Step six: search in other languages
Some inventions are better covered, explored, researched or filed in particular countries. The source gives an example: in China, an invention may have more literature than in the United States. In such a case, a Chinese native-language search is a sensible move, so prior art that could invalidate the patent isn't missed. Native Korean and Japanese searches are possible too. Search strings and hits are analyzed in different countries' databases and languages.
For Company B, this might be the difference between finding a decisive document and never knowing it existed. A document sitting in a foreign-language database is still public, and it can still count as prior art if it was available before the cut-off date.
From a Pile of Documents to a Persuasive Case: Mapping the Prior Art
By now the team may have a large bucket of results. Most of it won't be useful. The next stage is about picking out the few documents that really matter.
Narrowing down
It is critical to reduce the short-listed citations to the ones that come closest. The analyst assesses each citation against each claim feature. The citations that have most of the claim features get pulled up above the rest.
Each shortlisted citation is then examined carefully for its subject matter and its credibility. A document that looks perfect but cannot be shown to have been public before the cut-off date is of little use in court.
Then the analyst extracts the relevant text and presents it in a format that suits the client. There are two main ways to do this.
Summary type mapping
Here, relevant excerpts from a patent are provided. They may be highlighted in a color in line with the claim requirements. For example, the relevant text could be in red bold and the rest in black.
This format also includes a comment from the searcher, written in layman's language, giving a quick summary of what is and isn't present in the mapped citation. It can help the client work out whether the analyst inferred any equipment, keyword or similar element beyond what the claims required.
Claim chart mapping
In this format, the claims are broken into key elements. The relevant text for each key element is provided, and text for each feature is highlighted in a different color. The searcher adds comments for every key feature, so the client can understand each one separately.
This format is especially useful when more than one reference is being used. Say two references are combined. The missing elements show up clearly in the mapping of the first reference. The second reference, which covers those missing pieces, can be mapped right alongside. The client gets a clear, high-level view of how the two together cover the whole claim.
Which one to pick
Time is the deciding factor. When the project's duration is limited and the number of references to present is on the high side, the summary format is the better choice. Claim charts give a deeper understanding of a specific reference, but they take longer to prepare.
Company B, facing a court date, will likely want to talk this through with the team early, so the format fits both the deadline and the legal team's needs.
The Report, the Check-Ins and the Quality Gate
After the best citations are mapped, they go into a report. The report can be in Word, Excel or PowerPoint format.
A good report doesn't stop at the best finds. It also lists the closest possible citations and others that are broadly related to the invention, along with their bibliographic information. It may include a key feature analysis table too. That table shows which features of a mapped reference are covered, not covered, or inferred by the analyst. The client can scan it and judge the references quickly.
Different report styles for different needs
Reports come in several shapes, and the choice affects the budget. Options include:
A Word file with identified citations in summary form and the searcher's comment
A Word file with a thorough analysis of the results against the subject patent, with each key aspect mapped and commented on
An Excel spreadsheet in which each patent is summarized or mapped against each key aspect of the claims
Highlighted PDF copies of the identified citations, with the relevant text marked, which saves report-writing time so more time goes to searching
A table of relevant citations with their bibliographic information
The trade-off is straightforward. The less time a report takes to prepare, the lower the budget it needs. The more detailed it is, the more time and money it takes.
Interim reports keep you in the loop
Rather than going quiet for weeks and then dropping a final report, good providers send interim reports at regular intervals, followed by a final report. These tell the client what has been found at each stage and how the search is progressing. Based on them, the client can help steer the search toward their expectations. If Company B's legal team realizes a particular claim matters more than expected, the interim report is a chance to say so.
Quality control
Before anything goes out, the report is checked at several levels, including by a project lead and then the project manager. The check covers formatting and also the high-level discussion of the identified and mapped references. The aim is that only top-quality reports reach the client.
Talking to your search team
Regular phone and email conversations are encouraged throughout. A project usually starts with an introduction call or email to learn the client's expectations, requirements and search criteria. Interim reports are then discussed with the client regularly, or whenever the client needs. The search is steered based on the client's input. That keeps it on track and makes sure no part of the search gets missed.
What Will It Cost? The Honest Breakdown
Now to the question Company B's finance team is asking. The cost of invalidating a patent depends on several factors, and the source is clear that none of the numbers are fixed. They differ from person to person and from organization to organization. Think of the figures below as starting points, not quotes.
Who does the work
This is the first cost factor. Freelancers cost less than organizations, but search quality, standard and depth are affected. Organizations have teams of experts across many technology fields, which helps them understand the invention they are trying to invalidate and run a better search than one person can.
Organizations typically assign a group to a specific case. The group has searchers of varying experience and a team lead who oversees search quality and the references found, so nothing is overlooked. Others work on search quality, report quality and client interactions, so that no stone is left unturned.
On price, an individual may charge as little as $1,000 for this kind of search. Organizations typically start at $2,500. That is the starting point, and it may rise based on other factors.
How complex the invention is
Complexity is the second factor. Simple inventions in an easy domain, with fewer elements to search and smaller, simpler claims, can be done on a lower budget. Complex inventions in a hard domain, or with many claim elements to invalidate, need a larger one. They take more time to understand, to build search strings for, and to analyze. Larger search strings covering all the elements must be applied and analyzed to cover all possible prior art. The search may begin at $2,500 and rise with complexity.
Your budget, needs and the patent's history
Client budget, timeline, requirements and expectations form the third factor. Clients have varying budgets and requirements, and the cost is set based on those. Clients choose or are given a budget based on how complex the case is.
The patent's own history matters too. A patent that has been through multiple legal proceedings is challenging to invalidate. And the arguments and prior art disclosed by the examiner and the inventor set the scope of what remains to be found.
If a patent has few known prior arts, the searcher has a broader area to cover and more citations that may have been left out and could be used against it. If a patent has many backward citations and the examiner identified various prior arts, the searcher has many references to exclude, which makes invalidation harder. Two patents that look equally important on paper can therefore need very different amounts of work.
Non-patent and standards searches
Searching non-patent literature in addition to patents adds to the cost. These searches use databases such as Google, IEEE, Science Direct and Springer. A non-patent search could cost as little as $1,500. A standards search, on sources such as IETF, ETSI and 3GPP, may also start at $1,500.
Native-language searches
Searching in other languages is an extra cost factor. Providers with experts in various countries who are fluent in native languages can search local databases. A native search may begin at $1,500.
Report format
The report format you select also affects the price. A short summary with highlighted documents takes less time than a fully mapped claim chart. In addition to the search budget, a minimum budget of $200 is required for claim-charted reports.
Why the price makes sense
Organizations justify their prices by pointing to quality and subject matter expertise: experts in many fields and technologies across countries, and experts with research paper backgrounds who know the non-patent literature well enough to run high-level searches.
Put the numbers in context. The source says that when a lawsuit is rendered ineffective by invalidating the patent behind it, it could save a company millions of dollars. Against that, the cost of a careful search is a small part of the risk it is meant to address.
Who Should Run the Search for You?
Company B has a decision to make. Private firms hire professionals to run these searches. The firms are registered and have professionals with deep knowledge of the IP industry. The people doing the work are known as IP or patent researchers or analysts.
Before the search is carried out, a non-disclosure agreement (NDA) is signed for every search. For a company in the middle of a lawsuit, that protection matters a great deal.
Customers have options. The search could be carried out by an individual or an organization. An individual can offer limited expertise in a specific technology or IP field. An organization has a diverse workforce with wide IP experience and expertise across technology domains.
So the choice often comes down to what is at stake. If the case is simple and the budget is tight, an individual may be enough. If the invention is complex, or the lawsuit is serious, the depth of a team can be worth the higher price.
The source also describes how an organization would handle the project: a team assigned to the case, a lead supervising quality, regular communication, interim reports, and layered quality checks. These are the things to ask any provider about before you hire. A few questions worth putting to anyone you are considering:
Who will be on the team, and who checks the work?
Will you receive interim reports along the way?
Can they search non-patent literature, standards and other languages if the case needs it?
Which report formats do they offer, and how do those affect the price?
Will an NDA be signed before work begins?
When Should You Order a Search? Recognizing the Signals
Company B knew it needed a search because a lawsuit arrived. But the same logic applies earlier and in other situations. A search is a date-restricted prior art search that reveals the subject patent's technology. Its job is to challenge a claim that has already been granted, and to protect an entity involved in an infringement case by invalidating the claims of interest.
Here are the signals that a search might be useful for you:
A competitor has sued you, or threatened to, over a patent
You plan to launch a product that might infringe someone's claims
A competitor was granted a patent in your technology area, and you want a stronger hold on the field
You own patents and want to know how strong they are before enforcing them
You are thinking about licensing someone's invention
You are considering acquiring a company for its patents
The need for these searches exists because the examiner may miss potential prior art. And the search matters throughout the life of the patent, to determine whether a patent is novel or non-obvious during its term.
Two more hypothetical stories show how it plays out in other settings.
In the first, Company A, a patent holder, wants to extend its product or service portfolio. It wants to acquire or merge with a second company that holds a patented technology it likes. Before moving, Company A carries out a search on the second company's patents. It tries to validate the claims on novelty and non-obviousness. If the claims hold up, Company A can feel more confident investing in licensing opportunities with the other company. The search also tells Company A the strength and value of the patent, so it can assess commercial benefits and, if the patent is valid, monetize it through licensing.
In the second, Company B is a defendant protecting itself from monetary losses. It looks for prior art that reveals the claim limitations, both patent and non-patent, to invalidate the claims of infringement. Notice the mirror image: in one story the search is hoping to confirm a patent, and in the other it is hoping to break one.
The same skills and the same process serve both situations. What changes is the question being asked.
The Report in One Page: Putting It All Together
By the end, what does Company B actually hold in its hands?
The patent invalidity report, also called the invalidation search report or prior art search report, is an analysis of the references found during the search that can be used to invalidate a patent.
Here is how that comes together. The searcher has nearly used up the search budget and has run extensive patent and non-patent searches on the subject patent. They short-list a small number of citations that could be used against it. Each one is closely examined for subject matter and credibility.
To invalidate a patent, the invention has to fall under the laws of the relevant jurisdictions, such as US laws and EP laws. If the subject patent is from the United States, USC 102 must be considered when analyzing the prior art. Before any detailed mapping, the right parameters need to be confirmed.
If all goes well, Company B ends up with the following:
A short list of the strongest references
A mapping of those references against the claim features
A clear view of what is covered, not covered or inferred
A set of broadly related citations in reserve
A report in a format that suits the legal team's needs
Armed with that, Company B has real choices. It can argue in court that the patent should not stand. It can use the findings as the basis for a PTAB petition. Or it can enter settlement talks from a much stronger position. Finding good prior art does not guarantee a win, and a patent can be difficult to invalidate. But having evidence beats having none.
Wrapping Up: Turning a Scary Letter into a Plan
The letter that arrives on that Monday morning feels like a verdict. It isn't. A granted patent can be examined, tested and, if the evidence is there, challenged.
Here is what we covered. A patent invalidity search is a prior art search on a granted patent, aimed at showing its claims weren't novel, were obvious or lacked an inventive step. The same kind of search, run as a validity search, helps patent owners, licensors and buyers see how strong a patent is. The process starts with a team and a cut-off date, relies on a careful read of the patent and its file history, moves through citation analysis, keyword and class strings, non-patent literature, standards and native-language searches, and ends in mapped, quality-checked reports with regular check-ins along the way.
Costs vary. Individuals may charge as little as $1,000 and organizations typically start at $2,500, with non-patent, standards and native-language searches each starting around $1,500, and claim-charted reports needing a minimum of $200 on top. Complexity, the patent's history, your timeline and the report format all move the number.
If you are in Company B's shoes, or you are weighing a licensing deal, an acquisition or a challenge to a competitor's patent, the best time to start is before the pressure peaks. A clear view of the prior art turns uncertainty into a plan.
Ready to see what a focused search could uncover for your situation? Start with a Patent invalidity search: https://iebrain.com/patent-invalidity-search-a-complete-guideline/