Patents

A patent is a time-limited, government-granted right to stop others from making, using or selling an invention. It is published as a document with a unique publication number, such as US10123456B2, and its legally binding scope lives in its claims. Everything else in the document, the abstract, description and drawings, is there to explain and support those claims.

Patents are the raw material of a freedom-to-operate or infringement study. Before you can judge whether a product is clear to launch, or whether someone else's product practices an invention, you need the patents that could bear on it, read in a form you can actually work with.

This page is about how Patalyze represents a patent and how you read one. It covers the record we gather from the global corpus, the ways to search and add patents, a tour of the parts inside a patent, and how a patent connects to your products through a mapping.

What's the deal of patents#

A patent is a bargain with the state. In return for the exclusive right to commercialize an invention for up to twenty years, its owner has to publish how the invention works. The disclosure is the price of the monopoly, and the published document is the patent itself.

Publication is the whole point. A piece of knowledge only has to be worked out once, instead of several companies developing the same thing in secret and in parallel, and everyone else can read it and build on it, which is meant to move technical progress along faster. The right is deliberately not forever: it runs out after twenty years, and the renewal fees that keep it alive climb over time, so a patent is held only as long as it is worth paying for.

For the reasoning behind all this, see the post So why patents? Whats the deal?

How we understand patents#

Inside a research database, a patent is identified by its publication number alone, the unique code printed on every patent document, such as US10123456B2 or EP1234567A1. The full record does not live in your database.

Instead, Patalyze reads the patent the way an analyst would. Give it a publication number and it gathers everything known about that single published patent from a global corpus of millions of patents, then normalizes it so a US, European or Japanese filing all read the same way. The result is one record you can work with.

Four things anchor every record:

• Identity: the publication number, and the status that says whether the patent is still in force.
• What it protects: the title, the abstract and, at the legal core, the claims.
• Classification: a set of technology codes you can search and filter on.
• Where else it is filed: its family, the same invention published in other countries.

Everything else, the dates, applicants, inventors, figures and prosecution history, hangs off those four.

Search & add patents#

From the Add menu in a research database there are two ways to bring patents in: search the global corpus for the ones you do not know yet, or enter the numbers of patents you already have in hand.

Search for patents#

Search runs over the shared global corpus, the same millions of patents behind every research database and the Data API. A single query can work three ways at once, and a good search usually uses all three: it matches words, matches meaning, and narrows with filters.

Keyword search matches exact words and phrases in a patent's title, abstract and claims. Terms combine with AND by default, so each one has to appear. A trailing * stands in for any ending, so electrolyte* also catches electrolytes. It is precise, but it only finds the words you think to type.

Semantic search matches meaning instead of words. Describe the invention in plain language, or drop in a claim or a whole patent, and Patalyze ranks the corpus by how close each patent sits in meaning. It surfaces the filings that describe the same idea in different words, the ones a keyword would miss.

Filters narrow the field by the structured parts of a record. Each filter is a field, an operator and one or more values, and you stack as many as you need, combined with And or Or. The fields you can filter on:

• Text and Claims: words or meaning, across the whole record or the claims alone.
• Classifications: CPC and IPC codes, matched by prefix, so a broad code catches everything beneath it.
• Assignees and Inventors: who owns the patent and who is named on it.
• Territory and Kind code: the office a patent was published in, and the kind of publication (A1, B2).
• Status: whether a patent is in force, pending or dead.
• Priority date: before, after or between dates.
• Number: a specific publication or application number.

Put together, a freedom-to-operate search for solid-state battery electrolytes might read like this:

Every filter narrows the count, shown live as you build, and the matches preview in a table underneath. Swap the Text operator to similar to to rank by meaning instead of exact words. When the field looks right, add the patents that fit and they drop into your database.

Enter patent numbers#

Sometimes you already know exactly which patents you want: a competitor's filing a colleague flagged, or the patents named in an earlier report. Skip the search and enter their publication numbers directly, such as US10629950B2 or EP3324419B1. You can paste a whole list at once.

Either way, importing runs in the background. A patent's publication number appears right away, and the rest of its record fills in as the corpus is read. You can keep working while the claims and dates arrive.

Inside a patent#

Open a patent and it fills the page. Title and abstract sit up top, then the full claim tree, the figures, the family, a timeline of legal events and the patents it cites. Each part is its own block, rendered exactly as you see it in the app.

The tour below follows the order they matter in. We start with the claims, since they are the legal core, then move out to the family, classifications, legal events, register documents and citations.

Claims#

The claims are the patent: the numbered, legally binding definition of what it protects. A claim is a single sentence broken into elements, each one a feature that has to be present for the claim to read on a product.

Claim 1 stands on its own as an independent claim. The claims after it can depend on it, narrowing it by adding further elements.

That element-by-element structure is exactly what a mapping works on. To compare this patent against a product, Patalyze takes each element in turn and judges whether the product has it. That is why the claims are where any infringement read begins.

Families#

A single invention is rarely filed once. File it in the US, Europe, Japan, China and Korea and you get five publications for one underlying invention: a patent family.

That matters for risk. A competitor's patent filed across those five offices is one threat, not five, and the family ties the publications together so you can read them as one.

Each member carries its own status, because the same invention can be in force in one country and lapsed in another. For the wider picture, the patent family tree is a good primer.

Classifications#

Every patent is tagged with CPC and IPC classification codes that place it on a shared technology map, like H01M for a broad area or H01M 10/0562 for a narrow one. The codes are hierarchical, narrowing from a broad section down to a specific group.

You can search and filter on them to hold a technology landscape in focus, or to surface patents a keyword would miss because they never use your vocabulary.

If publication numbers themselves are new to you, the guide to patent numbers is a good place to start.

Legal events#

A patent's life is a sequence of events: granted, renewed, assigned to a new owner, opposed, lapsed, revived. Patalyze gathers the legal events from every office that has touched the patent into one timeline. At a glance you can see whether it is alive, who owns it now, and what has happened along the way.

Register documents#

For European patents, the EPO publishes the file: the documents exchanged during examination and opposition. Patalyze links these register documents straight from the record, so you can open the examiner's communication or an opponent's notice without leaving the page.

Citations#

A patent cites the prior art it was weighed against, and is cited in turn by later patents. The citations show what the examiner considered, each reference tagged by where it came up, in the search report, examination or opposition, and linked out to the cited document.

Description#

Behind the claims sits the description: the full specification, often tens of pages, where the invention is set out in detail. Patalyze keeps it a click away.

Open it in a side viewer next to the claims, its paragraphs numbered the way the patent office numbers them. Every reference to a figure is linked, so you can jump from a line of text to the drawing it describes.

The description is also where the claim language is pinned down. A term that reads as broad in a claim is often narrowed by the examples here, so reading it is part of judging how far a claim really reaches.

Figures#

Most patents carry drawings, and the drawings often carry detail the words skip. Patalyze shows them as a grid on the record and opens any one full-screen.

In the viewer it reads the figure for you. The reference numerals scattered across a drawing, the small numbers that label each part, are detected and turned into links back to the description. A number on the drawing takes you straight to the paragraph that explains it.

Click any thumbnail in the record above to open the viewer, then follow a numeral into the text. It turns a wall of unlabelled drawings into something you can actually navigate.

Mapping to products#

A patent only matters in relation to a product, whether that is your own offering under freedom-to-operate review or a competitor's that might practice your invention. A mapping is where the two meet. Patalyze takes each of the patent's claims, breaks it into the elements that must all be present, and judges the product against each one, with the evidence behind every call.

Each element gets a verdict, and the verdicts roll up into a score, so a wall of claim language becomes a result you can scan. Patalyze maps the patents in a database against your products automatically, and you can map any patent to a product by hand from the record. The full method, with the statuses, scores and risk bands, lives on the mappings page.

Working with a patent#

Beyond reading the record, a patent is a place to work. With one open you can:

• Read the source by opening the full PDF or the patent's description in a side viewer, without leaving the page.
• Ask about it: put a question to the patent directly and Patalyze answers from its text, handy when the claims run dense.
• Map it to a product, checking this patent's claims against one of your products on the spot, the same comparison the mappings page covers in full.
• Take it with you: copy a link to share the record, or export it to Excel with its full metadata.