Skip to main content

Premium patent data: what it is and how to use it

Summary

Premium assessments include a patent search. It finds patents and published applications that describe technology conceptually close to yours, and it compares meaning rather than legal claims.

Read the output as a query worth validating, not as a prior-art position. It gives you a strong starting list for a conversation with your patent attorney. It is not a search an attorney can put their name to without doing their own work.

The results appear on the evaluation covering IP viability, and only on Premium assessments. For what else changes between tiers, see Basic and Premium: what changes.

The patent search is a research tool. Nothing in the assessment or in this article is legal advice, and FirstIgnite makes no warranty as to the completeness or accuracy of search results. Decisions about filing, disclosure, or freedom to operate belong with your patent counsel.

When to use

  • You want an early read on how crowded a technical area looks, before you commit attorney hours to it.

  • You want documents a keyword search would miss because their authors used different vocabulary for the same idea.

  • You are assembling a list of candidate references to review with counsel.

  • You want to ask an inventor better questions about who else is working in the area.

Each of those is a question about where to look next. None of them is a legal conclusion, and the patent search cannot supply one.

Do NOT use the patent search to decide whether to file, whether you are clear to practice, or whether a disclosure obligation applies. Those decisions rest on claim-by-claim analysis and legal judgment, and this search performs neither.

How it works

What the search compares

The assessment reads the material you submitted and looks for patents and published applications that describe similar technology. The comparison rests on how closely the meaning matches, not on claim language and not on any legal test.

As a result, a document may rank highly because it addresses a similar problem using a similar approach, even if its claims cover a different subject. The reverse can also occur: a document with highly relevant claims may rank lower if it describes the same underlying idea using unfamiliar terminology.

Coverage is not universal. The search does not reach every patent office or every publication date, and very recent publications may not be included yet. Absence from the results is not evidence that a document does not exist.

What each result shows

Every result carries the patent number, title, owner, and date, plus a relevance figure. Opening a result shows the document's Claims text alongside an AI analysis of how the document compares to your technology. Each result also shows the passages the analysis judged to overlap with your technology.

The analysis quotes those passages directly rather than paraphrasing them, so you can check each one against the source text yourself. If a quoted passage does not match the source document, rely on the source. They show you the specific wording that made a document surface, which is the detail counsel needs to judge whether it matters.

Results also carry an indicated overlap level and a risk label. Both are automated readings of similarity, not legal conclusions, and the panel where they appear says so.

Patent references from an IP Viability evaluation. Each row carries a classification tag, the patent number and title, its assignee, and a line explaining how it bears on the technology.

Patent references from an IP Viability evaluation. Each row carries a classification tag, the patent number and title, its assignee, and a line explaining how it bears on the technology.

How to read the results

Treat every result as a candidate to verify, not as a finding.

Open the actual documents for anything you intend to act on. The analysis works from the text it receives. The claim view shows claim text only up to a length limit, and it says so on the page. A claim you need to read may sit past that limit.

Expect false positives. Conceptual similarity is not the same as relevance to your legal question, so some results will not survive a first read.

Expect gaps too. A document can be missing because its vocabulary differs from yours, or because it falls outside what the search covers.

Narrow the list as you read it. You can filter the results by concept, by owner, or by date, and you can sort them by relevance or recency. Select Flag on anything worth counsel's time, and dismiss the results you have ruled out. What you hand over then contains only what you want reviewed. Use Hide dismissed to keep the view clear.

What the patent search is not

  • It is not a patentability opinion.

  • It is not a prior-art search of record.

  • It is not a freedom-to-operate analysis.

  • It does not compare claims to claims, and it applies no legal standard.

  • It is not a search an attorney can rely on without conducting their own.

The relevance labels Blocking, Prior art, and Related rank technical similarity and an automated reading of freedom-to-operate risk. They are not legal determinations, they are not a materiality judgment under 37 CFR 1.56, and they do not decide what your attorney must disclose.

Nothing in the output establishes that your technology is patentable, or that you are clear to practice it. Those are legal questions, and they need someone qualified to answer them.

For how the IP-related evaluations turn this evidence into a score, see Evaluations: scope, limits and rubrics. For the other places an assessment gets its information, see Where the assessment's data comes from.

Tips

  • Read the quoted passages before you read the labels. The passages are verbatim text you can verify against the source, while the labels are automated readings of that text.

  • Do not compare relevance figures across assessments. A figure ranks a result within its own search only.

  • Send counsel the documents themselves rather than your summary of the results.

  • Filter by owner when you want to know who else is active in an area. The unfiltered list ranks results by similarity to your technology, which answers a different question.

  • If the results look sparse, raise it with counsel as an open question. Sparse results are not evidence that the area is clear.

  • Run a Premium assessment early, while the filing strategy is still open. The results are most useful when they can still change what you ask counsel to look at.

Common issues

The patent section says the data source was unavailable. The search could not run for that assessment, so the section is incomplete. An unavailable source is not a finding that no prior art exists, and you should not read the section as one. Run a new Premium assessment when you need the search.

No results appear. Either the search found nothing conceptually close, or the relevant documents fall outside its coverage. Neither case tells you the area is clear.

A result looks unrelated to your technology. That is expected. The search ranks conceptual similarity, and some of what it surfaces will not matter to your question. Dismiss it.

Your own patent is not in the results. When you submit a patent, the assessment excludes that patent from its own results, so it does not appear as prior art against itself. Related filings from the same patent family may still appear, as do other documents from the same owner.

A flagged passage has no highlight in the claim text. The passage can sit outside the claim excerpt shown on the panel, which covers only the start of the claims. Select USPTO to open the full published record at the patent office.

There is no patent section on your assessment. The patent search runs on Premium assessments only. An assessment's tier is fixed when you create it, so run a new Premium assessment rather than looking for an upgrade on the existing one. See Basic and Premium: what changes.

Frequently asked questions

Can I send these results to my patent attorney?

Yes, and that is what they are for. Send them as a candidate list to check, and say plainly that the search compared conceptual similarity rather than claims. Your attorney still has to run their own search.

Does a clean set of results mean my technology is patentable?

No. The search performs no patentability analysis, and its coverage is not universal. A sparse result set tells you where to look first and settles nothing on its own.

Does a result labeled Blocking mean I cannot practice the technology?

No. Blocking marks high technical similarity, which makes a document worth reading carefully. Whether it affects your freedom to operate is a legal question, and answering it takes the claim-by-claim analysis this search does not perform.

Can the results substitute for a prior-art search?

No. A professional conducts a prior-art search of record to a standard they are accountable for. This search surfaces conceptually close documents so you can tell your attorney where to look.

Why does the search find documents a keyword search misses?

It matches on the meaning of the text rather than on the exact words. That lets it surface a document that describes your idea in unfamiliar vocabulary. It is also why the search returns some results that turn out to be irrelevant.

Is the patent search available on Basic assessments?

No. It runs on Premium only. For everything else that differs, see Basic and Premium: what changes.

Did this answer your question?