“The important thing in science is not so much to obtain new facts as to discover new ways of thinking about them.” Sir William Bragg




    I was recently invited by a customer to discuss how or if they could reduce the number of captured patent documents per week from 200 to 5 (i.e. 40 times reduction) without risking missing out on relevant patents.

    In 2017 a total of nearly 3,168,900 patent applications were received by patent offices around the globe, which means that on average there were over 100.000 patent documents published every week on top of the almost 14 million patents currently in force according to WIPO. As a technology company, and even as a patent professional, it is overwhelming and very difficult to keep track of what is relevant for you without having efficient monitoring in place.

    Percentage shares of total filed patent applications in 2017

    Figure: Percentage shares of total filed patent applications in 2017 by the top five offices

    The workload for an R&D engineer or a scientist to go through and analyze so many patents per week will easily mount to many hours of not doing R&D activities. For a company with multiple R&D teams this can exceed by 1000 of hours per year. Therefore, the core question is: can this workload be reduced toa  few hours per week without missing important patents?

    Without reducing this workload, you are most probably also facing the risk of your R&D teams stop going through the documents and analysing them when week after week, they receive an email notification on a Monday morning that they need to analyze hundreds of new captured  patent documents of which only very few have some relevance to the R&D team’s field of work. I don’t know about you, but I would not like to start my Monday mornings like this.

    Email notification from the IAMIP platform

    Figure: Email notification from the IAMIP platform

    Faced with this, the customer concluded that they had two options; either stop monitoring third-party patents or create a patent monitoring algorithm that serves R&D engineers with fewer patent documents per week (e.g. 5). Both options carry a major risk of missing out on business-critical patents, that is why I was invited to their meeting to see what we could do to mitigate the risks.

    I asked my client to show me the search string that was used and the first thing I recognized was that they were using a combination of this

    3 methods to capture relevant data:

    Known competitors

    Keywords of the technical field

    Patent classifications

    The question now for me to answer was what creates the biggest noise in their weekly monitoring and based on my experience from the patent examiner carrier I could see that the part related to patent classifications in their string was most probably the source of the problem for them.

    Those of you who are not so familiar with classification-based searching can find more detailed information here. In nut-shell patent classifications are hierarchical codes defining a technical field that is assigned by patent examiners around the world throughout the examination phase based on the core of the invention and relevant field and is usually strongly connected to the patent claims. The patent applicant cannot affect the selection and therefore these cannot be extracted from the title or abstract in a patent application. Bear in mind that there are multiple patent classifications available such as IPC (International Patent Classifications), CPC (Co-operative patent classification), FI (File Index) etc. Therefore, depending on your technical field, competitor’s origin and so on you need, before composing your monitoring search string, to define which classification to use.  

    The advantage of using patent classification in monitoring is that they are:

    Independent of the language

    Independent of company internal wordings

    Independent of changes in terminology

    Assigned by authorized patent examiners

    But if you are using the classifications wrongly, they can create noise in your monitoring strategy as well as lead to you missing out on capturing relevant documents. Therefore, it is very important that you include classification in the right way.

    Three causes that can overwhelm your team with patent documents: 

    Using patent classification on the 1st level or 2nd level of hierarchy thus defining a very broad technical field.

    Including non-relevant patent classifications in your string.

    Not defining relevant regions and selecting a wrong classification system for that region

    Going back to my story about the client that wanted to reduce the number of patent documents by 40 times. The problem was that they used the IPC patent classification almost on the 1st level of the hierarchy which of course is a much broader definition of a certain technical field. In addition, one of its main markets was Germany. The German Patent Office does not classify patents with the CPC classification system, which is in many cases more granular and specific than IPC and thus allows for the creation of more precise search strings. But recently some patent database started to provide a mapping system that assigns CPCs to German patent documents and thus make them searchable in a more precise way.

    Personally, I have always preferred to use the CPC classification because I know that my former fellow colleagues from the European patent office and Swedish Patent office are trained to use this CPC classification system and therefore the relevancy is usually very good. But in the case of my customer, with their important market being Germany where patents were previously not classified in CPC, they avoided CPC and used IPC instead. Due to the recent change we concluded that, by using the CPC in combination with exclusion strategy (not relevant technical field) we managed to reduce the number of weekly captured new patent documents 40 times.

    The week after this, one of the email notifications that the R&D team received verified that we have managed to achieve the goal and to be honest everything below 10 makes you eager to see what are these about, right?

    The number of documents after improving the search string

    Figure: The number of documents after improving the search string

    If you as a search specialist can support your R&D team like this, you will become their internal rock star, because suddenly they only have to spend a few hours a week to review relevant patents and can focus the rest of their time on their passion of R&D activities.

    If you have some experience in how you reduce your monitoring results, feel free to share your thoughts.

    Are you also faced with a similar challenge reach out to us and one of our IP experts will support you with best practice? Book a meeting with our patent experts.

    One last thing, did you know that three of IAMIPs IP expert team are formers patent examiners?

    In this series of articles, I want to show you strategies to minimize this risk and monitor the patent situation in your technical area with simple and time-saving means. Become active instead of having to react defensively to the strategies of the competition!

    My name is Dimitris Giannoccaro, I am the CEO and co-founder of IAMIP and I am supporting technology companies to unleash the power in patents.

About the blog

The blog will share business insights based on our experience about how to manage intellectual property assets and rights in a smart way. We will also share best practice from working with the world's most innovative companies. Get to know IamIP better by following our key events from our entrepreneurial journey aiming at revolutionising the intellectual property world.

Latest IAMIP News

Here you can follow the latest news from IAMIP. At the bottom of the page you can register for our newsletter to get general news, webinar invitations and more.