How to Measure the Quality of a Patent – and a Patent Law Firm

All patents – and all patent applications – aren’t created equal. But how do you measure the quality of a patent or application?

As a white paper from the Canadian Intellectual Property Office notes,

It is often said that “you can’t manage what you can’t measure”. However, there is no real consensus as to how to measure the quality of a patent. Some IP professionals speak of number of words in the independent claims, the overall number of claims, ratio of independent claims to dependent claims, number of alternative embodiments, number of office actions sent before grant, and number of forward citations. We could come up with countless other metrics. The truth of it though, is that it is extremely difficult to determine a true metric relating to the quality of a patent. Most metrics suggested are highly qualitative or subjective.

One measure of the quality of a patent application is whether a patent was issued based on that application.

285,113 US utility patent applications from US entities were filed in 2019, and 157,315 were granted – 58.6%. (The correlation isn’t exact because most of the patent applications that resulted in patents in 2019 were filed in previous years.)

Many patents aren’t granted the first time an application is filed. According to the PatentRebel blog,

A study conducted by Yale University showed that between 1963 to 2005, the USPTO allowed only 11.4% of new patent applications that are unrelated to any other patent application while giving a non-final rejection to more than 86% of newly filed patent applications.

Another complication in calculating the percent of patent applications granted has to do with continuations. A “continuation” application is a new patent application that allows the applicant to pursue additional claims based on the same description and priority date as a pending “parent” patent application.

Patent applications are rejected for various reasons. The most common are:

• The invention is not “novel” (new).
• The invention is obvious in light of the prior art (previous inventions).
• Errors in the patent application

The most common types of application errors are:

• Mistakes in reference numbers on drawings
• Mistakes in paragraph numbers
• Mistakes in line numbers
• Mistakes in grammar and/or punctuation
• Failing to include an abstract (summary)
• Exceeding the authorized word count for the abstract

Patent applications can also be rejected because the patent claims aren’t drafted clearly, or for many other reasons.

PatentBots has developed a system for assigning Patent Quality Scores to evaluate and rank patent applications by law firms and patent owners. The score is based on the following criteria:

• Numbering errors
• Dependency errors
• Antecedent basis errors
• Word support errors

The company downloaded an entire year’s worth of more than 306,000 patent and proofread them with an automated patent proofreading tool. The tool then computed scores and rankings.

We’re proud to say that our firm ranked in the top 25 overall and in the top five among large firms with at least 500 issued patents. We also ranked in the top five for Tech Center 2800, which covers Semiconductors/Memory, Circuits/Measuring and Testing, Optics/Photocopying, and Printing/Measuring and Testing.