Patent landscape

WIPO describes its patent landscape reports as a snapshot of the patent situation of a specific technology, either within a given country or region or globally³. The Guidelines for Preparing Patent Landscape Reports (2015) say a landscape normally seeks to answer specific policy or practical questions, and that industry has long used landscapes to make strategic decisions on investments, R&D directions, competitors' activity and freedom to operate in introducing new products¹. They add that public policymakers increasingly turn to landscaping to build a factual foundation, especially in fields such as health, food security and the environment¹.

The Guidelines say producing a report can be a time-intensive and expensive process, and that an organization often does so when preparing to make a significant monetary or headcount investment in developing or moving into a technology area¹. WIPO designed the Guidelines both for general users of patent information and for those who produce reports².

1. Plan and set the scopeThe analysts and the client agree on why the work is being done, who will use the report and what it will cover, including the time period, ideally in a Terms of Reference¹. The Guidelines' scoping choices include country coverage, time period coverage, the patent family reduction method, and whether to include non-patent literature¹.
2. Design the search strategyThe search has to produce a query that provides the right data for the business question behind the report, and outputs in formats that can be analyzed in the next step. The databases searched should cover the countries within the scope¹.
3. Clean and prepare the dataThe Guidelines call patent data messy: misspellings in the applicant, assignee and inventor fields need to be corrected, and names grouped, for example to account for mergers and acquisitions, before statistical analysis¹. The set is then reduced by patent family or invention, and a manual review may be needed if too many irrelevant documents remain¹. Documents are also tagged with technology categories¹.
4. AnalyzeMost analyses count items in certain patent fields, using lists for one field at a time and tables or co-occurrence matrices for two¹. The Guidelines also list additional analyses such as citation and co-inventor networks and spatial concept maps¹.
5. Visualize and reportOnce the statistics are produced, often in a simple spreadsheet program, the output is visualized and added to the report. The Guidelines say a written report should be part of the deliverables¹.

The Guidelines say a number of standard analyses can be found in almost every report¹. They list these standard statistical analyses: number of families or inventions, number of national phase entries, number of granted patents, technology categories, office of first and second filing, top applicants or assignees, top inventors, and highly cited patents¹. Three of them are described below.

For the written report, the Guidelines suggest an executive summary, introduction, background on the technology and on patents, the objectives of the report, a methodology section, the analysis, additional resources and conclusions. The methodology covers the search strategy, data preprocessing, the analysis methods used, and issues and limitations. Most reports are 20 to 40 pages long in long form¹.

The Guidelines define several other reports that use patent information, for reference¹:

White space analysis is a term used for looking for gaps in patenting⁴. In a WIPO Magazine article (November 7, 2023), Lakshmi Supriya of WIPO's Technology and Innovation Support Division describes slicing patent data in different ways to find areas where patenting activity is relatively lower, the white space⁴. The article gives one worked method rather than a standard one: group the patents by product or application category, count the applications in each category to measure innovation intensity, record the filing years to measure recency, and plot intensity against recency on a chart. An area with many older patents suggests a mature technology, and an area with a few relatively newer patents suggests potential space to innovate⁴.

The article attaches cautions to this. It says patents are only one indicator of innovation and that many other forces could explain gaps in patenting, such as technical reasons why filings are low⁴. The Guidelines warn, for spatial concept maps, that distance between documents depends on their similarity to the rest of the collection, so guesses cannot generally be made about what sort of document might occupy an empty space on a map¹.

• Method choices change the numbers. The Guidelines say some family or invention reduction is critical, that using extended families can dramatically underrepresent an organization's investment in the United States and Japan, and that the reduction method should be stated in the methodology or limitations section and applied consistently¹.
• Data preparation matters. The Guidelines note that more time is usually spent preparing the data than analyzing it, and that preprocessing significantly affects the statistical values obtained¹.
• Recent filings are not yet visible. The Guidelines point to the 18-month delay in publication of patent documents in most jurisdictions, so very recent developments may only appear in non-patent literature¹. See Non-patent literature.