A guide to Patent & Innovation Benchmarking

In this article, I am trying to outline some ideas and thoughts that should be kept in mind while conducting benchmarking studies. It is very common in corporations to benchmark themselves against their peers in the industry. It could be for variety of reasons for example, measure R&D effectiveness, cost control, find strategic insights etc.

Before jumping on practical challenges, there are some basic concepts that should be kept in mind before conducting these studies.

1.  What kind of benchmarking you are going to conduct?

There are two types of benchmarking

A. Internal benchmarking: These types of benchmarking studies are conducted to see how various divisions are doing in terms of protecting their products, utilizing budgets, etc. There could be variety of parameters on which divisions can be benchmarked and companies may seek to find best practice within each division.

 B. Competitive benchmarking: These types of benchmarking studies are conducted to see how company is performing in its task environment.  These studies may provide valuable insights if conducted accurately. These studies could reveal pattern, best practice employed by other players in the industry, impact of patents filed by the company etc.

2. What is objective of the study?
For any benchmarking study, analyst must freeze the objective of his/her study. Who is going to be seeing the report? What is it that you want to find out? What you want to show to your end customer? And finally what is scope of the study?

I found following practical challenges pertaining to these studies:

Availability & accuracy of data: One of the biggest challenges in conducting benchmarking studies is availability of sanitized patent data. It is very important to clean the data before putting in a nice fancy presentation.  For example, in the end result one wants to benchmark two companies in a certain country say the UK. Though, the problem statement looks very generic but it does not have a straight forward solution. One must look very deep into the data to get exact figures. Analyst will face challenges in getting EP granted patents designated in the UK or granted but lapsed, add acquired assets, discount divested assets, check correct assignee of publication, remove duplicates, and so forth. Unfortunately none of the commercial database to my knowledge gives error free data. Only solution to this problem is manual analysis!

Lack of understanding:  In some cases a superficial understanding of the parameter, nature of task environment, and legal intricacies could also lead to unfruitful study. A researcher should make a thorough understanding of key variables involved in the study.

Overlooking other factors & misinterpretation: Analyst should not Oversell patents! It is undisputed fact that patents play a major role in business. However, that’s not all. It is not recommended to credit or discredit patent benchmarks for organization performance.  Conclusion from benchmarking should be derived after looking at market, regulatory, legal and other business factors.

Too much statistics: It’s always good to have statistical validation of benchmarking. However, researcher must not throw statistical jargons to make things look sane. Follow the KISS! Keep it simple and sober.

Handy Skills: 

In my view these skills could add value in the analysis and somewhat important for conducting these studies (in no particular order)
1. Strong Business Research & Understanding of business 

2. Understanding of Patent laws & Data
3. Advanced Microsoft Excel
4. Visualization techniques
5. Understanding of Managerial Issues

Possible Audience & Possible applications

Patent and innovation benchmarking could have wide variety of audience and it is important to customize the result based on audience. Strategic thinkers, lawyer, R&D heads, marketing could derive endless benefit from such exercise.

Note: It would be great if readers could also leave their comments and share best practices and make this write up more interesting.