Paris, 9 September 2021 – Following our previous article presenting the different methods of rating patent quality, this publication outlines the new quality assessment method developed by Regimbeau. This new solution uses strategic information available in patent databases to assist the evaluator with legal, technological, economic, and geo-temporal criteria.
A number of aspects are standardized by the patent system, including defining protection by a set of claims as well as assigning one or more classifications to all patent publications. However, the use of a patent can be very different depending on the technical field of the invention. This difference must be taken into account when assessing quality. This is why the first step of this new method consists of assigning a main technical field and up to two secondary technical fields to an invention.
To this end, the International Patent Classification (IPC) has been segmented to define more than 60 technical fields grouped in the following six categories:
These categories and technical fields draw on the segmentation used by the European Patent Office (EPO) in the annual “Patent Index” report.
The present method is based on the calculation of indicators (or metrics) for each of the defined technical fields. These indicators are statistics accessible by querying patent databases.
For each technical field, the following six metrics are determined:
(a) Patent activity, measured by a volume of publications over a 10-year period,
(b) Frequency of litigation and/or opposition (particularly in Europe and in the United States),
(c) Growth rate of the number of first publications of patent families, measured over a 10-year period,
(d) Annual average of “citing” patent families (patents that are subsequent to a patent and which cite that patent in their description or search report), excluding self-citations (i.e., “citing” patents with the same owner as the assessed patent), in the citation categories related to novelty or inventive step (X, Y, USC 102, and USC 103 codes). Due to limitations related to the volume of families in this type of search, this average is estimated on the basis of families published no more than five years prior to the evaluation, and only for families which include a PCT application and EP and US counterparts,
(e) Frequency of licensing agreements (particularly in Europe and in the United States), subject to registration in the online patent registers;
(f) Rate of extension of patent protection internationally, measured by a proportion of patent families comprising either a PCT application or a title in each of the IP5 offices (i.e., including EP, US, CN, JP, and KR family members).
When applying this method to a patent comprising one or more technical field(s), metrics (a) to (f) may be used to satisfy several criteria, as detailed below.
DIN 77100 and ÖNORM A6801 standards for the monetary evaluation of a patent specify that legal, technical, and economic influencing factors must be taken into account by the evaluator.
According to the present method, legal, technological, and economic aspects are each given a score from zero to ten, which makes it possible to obtain a first quality score.
A fourth influencing factor relating to geo-temporal aspects (geographic coverage and remaining term) is then considered.
The legal score is obtained on the basis of the following three criteria: patent validity, claim scope, and ability to determine infringement.
This score reflects the impact of a patent in its technical field(s) and allows potential breakthrough innovations to be highlighted.
As in most automated patent quality assessment methods, the present solution associates patent quality with the volume of “citing” patents, excluding self-citations. However, the aim here is not to simply count the number of citing documents, as it is, for example, common to cite a greater number of prior art documents in the United States than in Europe. We therefore only consider documents that have been cited by an examiner as relevant to the assessment of novelty or inventive step, and then calculate an annual average of citations that is compared with the average observed in the technical field(s) of the invention (d).
Other variables used in this section include the growth rate of the number of first publications in the technical field(s) (c), as well as the following indices: generality (dispersion of the classifications of “citing” patents) and radicalness (dispersion of the classifications of the “cited” patents, excluding those of the assessed patent). These indices were introduced in our previous article.
Another originality with respect to existing methods is that the ecological impact of an invention is taken into account, making it possible to obtain a bonus reflecting beneficial aspects for the environment of the assessed patent. This has been made possible with the EPO’s introduction of the Y02 CPC patent classification in 2013, which identifies inventions involving climate change mitigation technologies. By analyzing the classifications of a patent, one can calculate the ecological impact of an invention. This method thus values technologies that are eco-friendly, which is probably the greatest challenge of future innovations.
To reflect the potential economic impact of an invention, the frequency of licensing agreements (e) and the rate of extension of patent protection internationally (f) in the corresponding technical field(s) are considered.
The following two criteria influence the final quality score of the patent as multiplicative factors to the score calculated on the basis of legal, technological, and economic aspects:
These two indices allow us to obtain the final quality score of the assessed patent. Their inclusion as multiplicative factors gives considerable importance to geographical coverage and the remaining term at the time of the evaluation. After applying the method to several cases, we have observed that the present method is more likely to generate a higher standard deviation of quality scores than existing “manual” methods due to its calculation process. This has the advantage of clearly characterizing patents with high potential.
The amount of time required for the calculation of the quality score ranges from about thirty minutes to one hour per patent. This means that it is possible to perform a quality assessment of a portfolio comprising a few dozen patents relatively quickly.
This new method of assessing patent quality includes both criteria requiring human analysis (for example, evaluation of claim scope and the degree of monitoring for potential infringement of the rights conferred) and criteria evaluated in an automated manner based on patent classifications, citations, geographical coverage, and remaining patent term.
One of the innovative aspects of this method as compared to existing solutions is the consideration of the ecological impact of an invention as part of the evaluation of its technological impact. This is not the only advantage, however, of the new method over existing methods, as new statistics are generated from the patent databases. Moreover, promising innovations in terms of technological breakthroughs are more efficiently detected and the time required for the evaluation makes the method perfectly adapted to patent portfolios comprising a few dozen patent families, such as those of many SMEs in France (see our “360° Audit” offer).
The method presented in this article makes it possible to assess the potential value of a patent or patent application. It is an evaluation of intrinsic quality rather than a financial evaluation that would require other criteria to be taken into consideration. Nevertheless, the quality score obtained here can advantageously be used as a reliable strategic indicator in portfolio reviews in order to identify potential opportunities to acquire or sell IP rights. Our teams remain at your disposal, should you need assistance in these matters.
