Stanford NPE Litigation Dataset

About Us

The Stanford Non-Practicing Entity (NPE) Litigation Dataset (the Dataset) is the first ever publicly available database to track comprehensively how practicing entities, non-practicing entities (NPEs), and patent assertion entities (PAEs) claim patent ownership rights in litigation. NPEs do not make products or offer services while PAEs—often referred to as “patent trolls”—employ patents primarily to obtain license fees, rather than support the transfer or commercialization of technology. Critics have come to believe that steadily increasing PAE enforcement activity, including litigation, is harming innovation and serving as a tax on producers and consumers. Stanford Law student researchers are tracking every lawsuit filed in U.S. district courts from 2000 to the present and identifying each patent plaintiff as either a practicing entity or as one of eleven types of NPEs. Categorization of all 55,000 lawsuits filed from 2000 to 2015 becomes publicly available in late 2017 and this full Dataset will continue to be updated with recent and future cases. The currently available compilation represents a 20% random sample of over 10,800 lawsuits filed from 2000 to 2015 and is available for download below.

Stanford Law student researchers and Stanford IP Research Fellow Shawn Miller have coauthored the Introduction to the Stanford NPE Litigation Dataset (the project report). The project report includes detailed explanation of the motivations for creating the Dataset, the lawsuit categorization taxonomy and methodology, and quality control measures taken in constructing the Dataset.

By the end of 2017, the project report will also include descriptive statistics and trends in the share of U.S. patent litigation attributable to different types of patent owners. For now we report that the 20% random sample reveals differences in patent litigation across each type of patent owner over the 16-year period, with a 3% margin of error. Key findings of the characteristics of patent plaintiffs include:

  1. increased PAE litigation,
  2. decreased practicing entity litigation,
  3. nearly 80% of software litigation is PAE litigation,
  4. suits involving PAEs terminated faster than other entity suits, and
  5. the win rate for practicing entities is higher than that of PAEs.

As it becomes available, scholarship using the data will appear on this website.

Brief Dataset Methodology

table of total lawsuits by patent asserter category
Table of total lawsuits by patent asserter category

As detailed in the project report, each patent asserter involved in a lawsuit has been assigned to one of 13 categories (see Table 1) and coded into the Dataset via a user interface created specifically for this project by Lex Machina. Each data point represents a lawsuit; some lawsuits contain multiple patent asserters and some of these multiple-asserter cases contain different types of patent owners. Such cases are assigned to multiple patent asserter categories.

Coders consulted several different sources to determine how to categorize each entity. First, coders reviewed Lex Machina’s listing for each lawsuit, then they read key pleadings including the initial complaint. Coders utilized Lex Machina’s links to the patents asserted to check the Patent Office’s assignments database to obtain clues about the current owner type. Coders then conducted a comprehensive web search for each unknown patent asserter. After determining the proper categories for the patent asserters in each lawsuit, coders assigned the numbers for those categories to the patent asserters via the user interface.

Note: In our Key Findings, we define PAE lawsuits as those with category 1, 4 or 5 patent asserters. See Table 1 for the name of each patent asserter category.


Request NPE Dataset

Fill out this form to receive a copy of the dataset.

Note: Do not use this form to submit High Risk Data.

Suggested Research Uses

Photo of light bulb with TRADEMARK conceptual words isolated on white

The Dataset can aid Congress and the Patent Office in developing effective laws and policies and provides a valuable tool to help patent holders, policy makers, judges, litigators, industry, scholars, and researchers better understand the nature of entities filing patent suits.By lending transparency to the amount and characteristics of PAE litigation, the Dataset can help resolve the fact or fiction of alleged PAE problems. Scholars, legal professionals, policy makers, patent holders, and the public will benefit from better understanding that may, in turn, yield more effective policies, laws, and practices. Scholars can also use the Dataset to examine the effects of past patent reform actions, such as the AIA, on different types of patent owners.

Scholars

  • Empirically confirm past research conclusions using the full population of lawsuits
  • Conduct original research on differences in enforcement behavior across patent owner type

Judges and Patent Attorneys

  • Gain understanding of patent litigation trends
  • Gain transparency into different types of patent asserters
  • Improve case management practices

Policy Makers

  • Develop a deeper understanding of the sources of patent enforcement costs given current rules
  • Enhance and develop regulations and statutes

Sponsors and Partner

This project is made possible through the generosity of the Charles Koch Foundation and Lex Machina. The research is grounded in a four-year study by the Stanford Law and Policy Lab “Empirical Study of Patent Troll Litigation” Practicum.

Charles Koch Foundation

Lex Machina logo
Law and Policy Lab