The 1790 Analytics Patent Scorecards provide a unique view into the leading innovators across major industries. The scorecards are based on objective, quantitative benchmarking of patent portfolios associated with more than 8,000 commercial enterprises, academic institutions, nonprofit organizations, and government agencies worldwide.
The scorecards take into account not only the size of organizations’ patent portfolios, but also their quality – as reflected in characteristics such as growth, impact, originality, and general applicability. The focus on both patent quantity and quality enables organizations with smaller, high-quality portfolios to fare well against competitors with much larger portfolios.
Individual industry scorecards may be accessed via the links below. All scorecards are also available in a single location via the All Scorecards link.
Details of how the scorecards are constructed can be found here: Constructing the 1790 Patent Scorecards.
Constructing the Patent Scorecards
Organizations in the scorecards are ranked based on their Pipeline Power, the final column in the scorecards. This is an overall measure of the strength of each organization’s patent portfolio, taking into account both its size and quality. The Pipeline Power for the year 2019 is calculated using the formula:
Pipeline Power = (Number of 2019 U.S. Patents) x (Pipeline Growth Index) x (Adjusted Pipeline Impact) x (Pipeline Originality) x (Pipeline Generality)
As this formula suggests, the Pipeline Power is derived by taking the extent of an organization’s recent patent activity. This figure is then weighted based on a number of other metrics that reflect the growth, impact, originality, and generality of the organization’s patent portfolio.
Number of U.S. Patents – This is the number of U.S. patents granted to an organization in the scorecard year. The scorecards are restricted to organizations with at least 100 granted U.S. patents in the selected year.
Pipeline Growth Index: This metric measures whether an organization’s patent activity is increasing (suggesting a strong ongoing commitment to innovation) or decreasing (suggesting a declining commitment to innovation). It is calculated by taking the number of U.S. patents granted to an organization in the most recent year (2019 in this case), and dividing it by the mean number of patents granted to the company per year for the previous five years (2014–2018).
Pipeline Growth values greater than 1.0 reflect an increase in patent activity (for example, a value of 1.3 shows a 30 percent increase in patenting in 2019, compared with the annual mean for 2014–2018). Meanwhile, values less than 1.0 reflect a decrease in patent activity (for example, a value of 0.8 shows a decrease of 20 percent in 2019 patenting relative to the annual mean for 2014–2018). The Pipeline Growth metric is capped at 2.0. Otherwise, organizations with few patents prior to 2019 could have extremely high Pipeline Growth values.
Adjusted Pipeline Impact: This metric shows the impact of an organization’s patent portfolio on subsequent technological developments. It is based on the idea that patents containing important technological information will form the basis for many new innovations, and so will tend to be cited as prior art by many later patents. This does not mean that every frequently cited patent is important, or that patents cited infrequently are necessarily trivial. However, numerous validation studies have revealed the existence of a strong positive relationship between citations and measures of technological importance.
The basic (unadjusted) Pipeline Impact metric is calculated by taking all patents granted in the most recent year and counting how many times they cite as prior art the patents granted to a particular organization over the previous five years. For example, the 2019 Pipeline Impact for IBM is derived by counting the number of times patents granted to IBM in 2014–2018 are cited as prior art by all patents issued in 2019. This number is then normalized by dividing it by the mean number of citations received by all patents from the same age and technology (as defined by patent classifications) as the IBM patents.
This normalization results in an index (the Pipeline Impact) with an expected value of 1.0. Pipeline Impact values above 1.0 show that an organization’s patents have been cited more frequently than expected (e.g. a value of 1.5 shows 50 percent more citations than peer patents) and are thus high impact on average. Meanwhile, Pipeline Impact values below 1.0 show that an organization’s patents have been cited less frequently than expected (e.g. a value of 0.7 shows 30 percent fewer citations than peer patents) and are thus low impact on average.
It is possible for the Pipeline Impact to be skewed by organizations citing their own earlier patents excessively as prior art. A certain number of Self-Citations is to be expected, and this is healthy, as organizations build on their own earlier research. However, extreme self-citation, combined with few citations from the patents of other organizations, may be an artifact of an organization’s patent filing process, rather than a reflection of technological impact. To account for this self-citation effect, we discount all self-citations that are more than 30 percent of the total. For example, if an organization has a Pipeline Impact of 1.2 and a self-citation rate of 45 percent (that is, 45 percent of the citations of the organization’s patents are from the organization’s own subsequent patents), we adjust its Pipeline Impact by 15 percent to 1.02. This modified figure is the Adjusted Pipeline Impact for the organization (capped at 5.0). This metric is included in the formula used to derive the Pipeline Power for an organization.
Pipeline Originality: This metric gives a higher score to patents whose prior art comes from a variety of technologies (as defined by patent classifications), rather than from the same narrow technology. It is based on the idea that incremental inventions tend to reference prior art from a single technology, whereas radical new inventions often combine two or more technologies to create a new patent. Like Pipeline Growth and Pipeline Impact, Pipeline Originality is normalized by year and technology and has an expected value of 1.0. Organizations with a Pipeline Originality value greater than 1.0 have patents that tend to reference a broad range of prior art and are thus regarded as more likely to be original. Organizations with a Pipeline Originality value less than 1.0, meanwhile, have patents that tend to reference a narrower range of prior art and hence may be less original.
Pipeline Generality: Pipeline Generality is similar to Pipeline Originality but is based on subsequent citing patents, rather than prior art references. Specifically, patents cited by subsequent patents from numerous technologies (as defined by patent classifications) are regarded as more general than those that are cited only by patents from a single technology. Pipeline Generality is again normalized by age and technology and has an expected value of 1.0. Values of greater than 1.0 suggest that an organization’s patents are more generally applicable than peer patents, while values less than 1.0 suggest that an organization’s patents have relatively narrow applicability relative to their peers. Pipeline Generality is capped at 5.0.