How to understand patent documents: Tips for non-experts

Imagine you are a scientist or engineer, and the head of R&D at your company has asked you to review some patent documentation. To someone unfamiliar with patents, the claim language can be confusing, the document structure is impenetrable, and the terminology used is obscure. Perhaps the patent is in your field of technical expertise, or an associated one. Likely, you have a special expertise to bring to bear, and the IP experts need to draw on your insights to interpret the precision and recall of their use case.

In fact, R&D engineers or scientists are frequently asked to review patent documents, sometimes to determine their relevance to their own company’s projects, or simply to better understand the patent environment in their technology space. Often, it’s to inform a “go/no go” decision – is there “freedom to operate” without infringing other patents. Perhaps it’s to access the patents of a potential licensing or acquisition target, or to vet the work of a scientist they might want to hire. In the Mind+Machine world of business analytics, in this case it is a scientist or engineer’s ‘mind’ that is needed. However, care needs to be taken to avoid unnecessary risks based on an incorrect interpretation of the data.

Whatever the reason, when R&D associates follow these tips, this tricky job becomes a piece of cake (or at least a little easier and less risky)!

When reviewing US patent documents, start with the icing!

The first thing you need to consider when reviewing US patent documents is the first page – think of this as the first, eye-catching (or distracting?) layer of icing on the “patent cake.” From this you can already tell:

The patent title. This can be succinct, detailed, or somewhere in between. Never base your review of a patent on its title alone, they can be deceptively vague. You can have an excessively wordy title like this or one like this titled “an automobile” which on closer inspection has absolutely nothing to do with a vehicle. An ideal review of a patent would consider the title, abstract and claims of the patent.

Bibliographic data

  • Inventor: This is useful to track their published research in the scientific literature.
  • Assignee: Keep in mind the assignee (owner) may have changed.
  • Legal status: A patent whose assignee (owner) has paid its maintenance fees may enforce that patent in the country in which it was granted, but the patent almost never discloses its legal status. Instead, this information can be found in subscription patent-search databases. If the patent is not in force, its primary value is as disqualifying prior art to prevent future applicants from patenting something identical or very similar.

References cited: These can be very helpful for identifying similar technologies that may be of interest to a researcher, e.g. in the publications cited by the inventor/applicant to support their invention’s patentability. Despite this, a scientist/engineer shouldn’t blindly trust references given – perhaps much time has passed since the application was filed, or the inventor/applicant just missed a reference.

Abstract: This is the 150-word description (or less) of the invention provided by the inventor. This can also be deceptively vague or can add valuable sources of information. If time is short, scientists asked to review a patent document should start their review on the abstract to find the most concise, focused, and readable description of the invention anywhere in the document. NB: Inventors walk a line between disclosing enough information to allow the patent examiner to determine the invention’s patentability, without disclosing information that lets a competitor easily make or reproduce the invention.

Drawing(s): Most patents contain detailed drawings to help the examiner understand what the invention entails. These can be incredibly useful to aid understanding of the invention. Unlike the textual components of the patent, remember that these can’t be searched, so they only serve to supplement the description in the patent.

Biting into the patent cake – the devil can be in the detail

Once you get past page one, digging into the patent cake itself can be even more satisfying – if more time consuming – and needing a bit more skill to find the really tasty bits:


The specification gives the background of the invention. Here the inventor/applicant makes the case in writing that their invention is patentable relative to everything that came before it, i.e. the “prior art”. It’s common for an applicant to craft an extremely broad background to obfuscate prior art searches, so researchers who aren’t familiar with the technology could be confused, e.g. when a patent application in telecommunications begins by referencing Marconi. In this case, the scientist should read the text and not get distracted by loosely related information included.

Relative to the claims, the detailed description lists the important elements of the invention and how they operate together with less “patentese” than in the claims (see below). Generally, this detailed description resembles a true technical expert’s description of their invention to a peer (rather than a patent attorney’s!).

The specification is where any new words employed by the inventor/applicant to describe some or all their invention will be found (see below).

Consider having an engineer/scientist read this section as a transition between the background (which will be more engineer/scientist-friendly) and the claims (which will be more patent expert friendly), and to keep the concepts discussed in the Claims section below in mind when reading it, in order to make it more understandable.


The claims are the boundaries of an invention. They contain the most critical elements of the invention from a legal perspective, not necessarily a technical perspective, and so any elements of the invention that are not listed may be fair game to be patented by others.

There is a concept in patent drafting that “the inventor is their own lexicographer,” meaning that the applicant can create new words in the course of describing their invention and use those words in the claims of the application, as long as those terms are clearly defined in the specification of the patent. This means the claims can be written in a language that will make no sense at first glance to a researcher (or even be contrary to the ordinary definition of the term) until they review the definition in the specification. Therefore, it is critical for an engineer/scientist reviewing a patent to understand this principle and not just assume the meaning of a word they’ve never seen before in the claims of a patent. 

Claims are typically formatted as independent and dependent, meaning that the independent claims are central components of the invention, and the dependent claims are patentable iterations of those central components that the inventor/applicant wants to include in the scope of their invention. An engineer/scientist should also keep in mind when reading the claims that each dependent claim is not conjunctive with the other dependent claims but disjunctive.

In conclusion, whichever part of the delicious patent cake you decide to ingest first (the summary “icing” on the first page, or the more detailed “filling” afterwards) – based on the special insights of a scientist or engineer with relevant technical knowledge, you will probably notice something tasty and interesting that the IP experts wouldn’t have been able to interpret alone. Just like Afternoon Tea – interpreting patent documentation is best done with a few friends for the best effect!


Disclaimer: the commentary in this blog post should not be construed as legal advice or legal opinion.

Share on facebook
Share on linkedin
Share on twitter
Share on email
Michael Fuller
Michael Fuller
Client Executive, IP and R&D Solutions Posts

Michael Fuller is a Client Executive, IP and R&D Solutions at Evalueserve. He is responsible for the East coast region of the United States, from Virginia to New Hampshire. Michael strives to be a trusted advisor, helping clients solve their most business-critical problems. A U.S. patent attorney (reg. no. 59606) himself, Michael has authored or co-authored over 40 legal and scientific articles, abstracts, and blog posts; has spoken on the topics of open innovation and patent analytics; and previously worked in pharmaceutical R&D on treatments for neurodegenerative diseases such as Parkinson's disease and Huntington's disease. He considers his role to be similar to that of a lawyer – as both have the opportunity to act as a fiduciary to client, and to support them in an informed way on difficult and critical subjects. He’s passionate about sharing knowledge and experience with different industries and those who might benefit from it!

Latest Posts