The world of patent law is highly opaque to most observers. To have authority in this world, not only must one be a member of legal priesthood, one must be a member of the engineering priesthood as well. This high barrier to entry makes it very difficult for clients to make data-driven distinctions between patent law firms. Firms are traditionally selected by reputation or based on relationships and the quality of their work rarely enters the equation. IP Street's Patent Claim Scope algorithm enables users to transcend this word-of-mouth analysis by providing measurable quality metrics for a patent law firm's actual work product, the claim text.
How IP Street's Claim Scope Scoring Works
Our claim scope algorithm compares every independent claim of every patent against all other independent claims in the given patent's class. Every patent claim is scored between 0% and 100%, with 100% being the broadest claim relative to its patent class and 0% being the narrowest.
The algorithm is based on two core premises:
- Claims with more terms relative to their class tend to be narrower. For example, "A buggy whip" and "A leather buggy whip" would both be members of US Patent Class 231: Whips and Whip Apparatus yet the former claim is broader than the latter. The inclusion of the additional word "leather" adds a limitation on the material of manufacture whereas the former claim covers buggy whips made of any material, say nylon or cotton.
- Claims with more unique words relative to their class tend to be narrower. Continuing with our example above, "A leather buggy whip" is likely to be a broader claim than "A WiFi-enabled buggy whip". Even though the two claims have the same number of terms, "leather" is much more common in US Patent Class 231: Whips and Whip Apparatus than the term "Wifi-enabled" and thus constitutes a more significant limitation. Notice that the inverse may be true if these claims instead lived in US Patent Class 701: Data Processing- Vehicle Navigation.
Claim Scope can be used to directly compare law firms against each other. Let's say you are an autonomous vehicle company looking to hire a new law firm to prosecute patents on your new GPS-enabled location sensing product. Since your patents are likely to fall into US Class 701: Data Processing- Vehicle Navigation, you should seek out a law firm that excels in this technology field.
Which firm writes the best vehicle navigation patents?
Querying IP Street's Basic Data Feed API endpoint is an easy way to request all the patents ever applied for in the class and returns about 45,500 US patent assets. A little slicing and dicing reveals that the top 5 law firm who have written the most patent applications in US Class 701 are Kenyon & Kenyon, Oliff & Berriage, Oblon Spivak, Sughrue Mion, and Crowell & Morning. Since each law firm has written 500-800 patents in this technology area, it's difficult to differentiate between the firms on the basis of quantity. They all have about the same level of experience as a firm. How about comparing these firms by the quality of the patents they produce? The IP Street Claim Scope algorithm let's you do just that.
Below is a side-by-side comparison of each of the top 5 law firms by Claim Scope score. Every patent they have written has been bucketed into one of four quartiles by their claim scope score. The red bars represent the number of patents that a firm has written which fall into the narrowest bucket. Similarly, the green bars the patents which fall into the broadest category.
There are statistically significant differences in the distributions. These differences show up both in the distribution chart above as well as the mean grant scope values of each firm.
|Firm||Mean Grant Scope Score|
|Kenyon & Kenyon||54.75%|
|Oliff & Berridge||54.59%|
|Crowell & Moring||49.87%|
Oliff & Berridge looks particularly strong in this technology area. Not only is their overall average relatively high, their overall distribution of assets is skewed to the right.
While there are many possible justifications, for their weakness, Sughrue Mion appears to write much narrower patents than the other 4 firms. They have the lowest mean score and their distribution is skewed significantly to the left. This difference is statistically significant at the 99% level.
As a quality minded consumer of patent prosecution services, Kenyon & Kenyon or Oliff & Berridge seem like the best bet. The claims they get allowed from the USPTO are significantly broader on average than other firms with similar experience in the technology class.
If you are interested in learning more about how IP Street's Claim Scope algorithm or one of our other patent analytics APIs can help you to better analyze patent claim quality, please sign up for a free trial subscription.