Patent Analytics – A Question of Timing

The amount of patent application, filed in various authorities around the world is growing every year at a dizzying pace. This vast amount of data included in the patent application enables the production of scientific, technological, business and competitive information. This information can be translated to varied insights, by analyzing and statistically processing this patent data and the characteristics that accompany each of the patent application records.

More specifically, through precise and responsible analysis of published patent applications, various trends can be identified by analyzing the application fields, filing countries and years of filing. Analysis can present the leading filing applicants, identify collaborations between varied institutes/companies, and even assist in identifying the patent attorneys’ fields of expertise.

The analysis of patent application data also enables varied comparative measures. For example, one can compare the total number of applications to the level of applications actually granted, thereby determining the quality of the applications and the different applicants IP.

Today, most of the commercial patent databases include built-in analytics tools. The results of the analysis can be displayed graphically, multi-dimensionally, and clearly. However, is all that the glitters gold? If the analysis is performed on incomplete data or is based on a mistaken understanding of the patent details, although the results might be visually impressive, they may also be fundamentally misleading.

Patent analysis requires an ability to make a professional patent search and find all the relevant publications on which the analysis will be performed, as well as deep understanding both of the raw patent data and of the technical tools. For example: Is the analysis performed based on families, on applications, or on granted patents? Is the analysis performed based on the priority date of the application, publication date, or date of grant? Does the analysis of applicants take into account reassignment?

One important point to consider when analyzing patent information is that the publication of patent applications takes place only 18 months after their filing. Sustainable and responsible analysis, which examines publications by year, must therefore be based on a full year in which all patent applications in that same year have already been published.

For example – analysis conducted in January 2018 will present only applications filed through June 2016. In addition, it will not be possible to relate to applications that were filed subsequently. Accordingly, in January 2018, one can only present a yearly analysis that relates to patent applications filed 2015.

Only authorities and entities like WIPO possess the data on the actual number of applications and can perform annual analyses and summaries as of the date of analysis. See for example: Patent Cooperation Treaty Yearly Review 2017

Recently, Managing Intellectual Property (MIP) published a report, which ranks the top patent-filing firms in each country according to the number of PCT applications for the years: 2014, 2015, and 2016. Since 2016 applications had not yet been published in full when making the report (18 months had not yet passed), the inclusion of a 2016 ranking creates an incomplete view of the data. In fact, only in July 2018 will it be possible to analyze all the publications of the year 2016.

In summary, patent application data analysis allows the creation of meaningful information from which one can reach unique insights that cannot be attained by other means and sources. In performing an analysis of this data, we need to understand the nature of the database and the meaning of the various parameters in order to use them wisely and carefully, so that the results will indeed accurately reflect the state of the subject.