PlainPatent

Editorial & Corrections Policy

Last updated:

This page explains where PlainPatent's information comes from, how it is produced, and how we handle corrections. We publish it because you should be able to judge for yourself how much to trust what you read here — and hold us to it.

Our source

Every patent figure on PlainPatent — patent counts, grant dates, assignee (company) names, CPC technology classifications, and claim counts — comes from a single official source: the USPTO PatentsView dataset, the research-grade patent database maintained by the U.S. Patent and Trademark Office. We do not scrape the web, accept user-submitted figures, or estimate values to fill gaps. Where a number is derived (for example, our composite innovation score), it is computed only from PatentsView fields, and the calculation is documented in our methodology and innovation-score explainers.

How pages are produced

PlainPatent is a data-journalism site, not a hand-written one. Company profiles, technology landscapes, and rankings are generated by a documented data pipeline:

  1. Load. We download the bulk PatentsView release and parse the patent, assignee, and CPC files into a structured database. Base counts and dates are loaded directly — never edited by hand.
  2. Compute. We aggregate patents into company-level profiles and compute derived metrics (technology breadth, filing velocity, claim depth, and the 0–100 innovation score) using the published weighting.
  3. Render. Each page reads those figures live from the database, so what you see reflects the loaded dataset rather than a cached copy of a number typed into an article.
  4. Frame. The editorial work is in the framing — the explainers, the plain-language context, the methodology, and the guides. That layer is written and maintained by the PlainPatent editorial team, operating under Kiznis Studio.

We are deliberate about the line between the two: we never alter an underlying patent figure to make a sentence read better, and we label plain-language explanation where it could be mistaken for an official USPTO statement.

Sourcing standards

  • Patent data is attributed to USPTO PatentsView, with the dataset vintage shown near the figures and in the sitewide data-sources note.
  • Explainers and guides cite primary or authoritative sources for load-bearing claims, with links you can follow.
  • We present patent volume, velocity, breadth, and claim depth as neutral metrics. We do not rate companies as "good" or "bad" innovators, and we do not editorialize a company up or down.
  • We take no payment, sponsorship, or promoted placement from the companies or entities we cover. Our only revenue is contextual display advertising, which has no influence on what we cover or how we present it.

Update cadence

USPTO PatentsView publishes updated bulk data on roughly a quarterly basis, with an inherent lag of several months between a patent's grant date and its appearance in the dataset. We refresh PlainPatent within 30 days of each new PatentsView release and update the vintage shown on the site when we do. The current dataset and coverage window are stated on our about and methodology pages.

Corrections process

If you believe a figure on PlainPatent is wrong, please tell us. Email hello@plainpatent.com (or use the contact form) and include:

  • the exact URL of the affected page;
  • the specific figure or sentence you believe is wrong;
  • what the correct value should be; and
  • a link or citation to the official source, where you have one.

Here is what happens next. We trace the figure back to its source field in the PatentsView data. If the value on PlainPatent does not match the source, that is a defect on our side: we fix it at the data or pipeline level so the correction holds through the next refresh, rather than patching a single page by hand. If the figure on PlainPatent does match PatentsView but the upstream record itself looks wrong, we cannot rewrite the official record — but we can add a correction note to the page and, where appropriate, flag it to PatentsView. Either way we will reply to let you know what we found.

We do not remove accurate information simply because it is unflattering to a company. We will consider removing or amending content only where it is materially inaccurate, outdated past its useful horizon, or subject to a legal right under applicable law.

Limitations we are upfront about

PlainPatent tracks US utility patents only — design patents, plant patents, and foreign patents are not included, so companies with large non-US portfolios can look underrepresented. Patent data reflects grants, not current R&D, and there is typically a multi-year lag between application and grant. Disambiguation is imperfect: PatentsView's machine-learning models occasionally split or merge company names, especially across subsidiaries and mergers. And patent counts measure R&D output, not commercial value. We say this plainly on every relevant page rather than burying it. For any legal or business decision, verify patent information against official USPTO records.

Who is responsible

PlainPatent is published by Kiznis Studio. The PlainPatent editorial team is accountable for these standards, for the framing and methodology, and for acting on the corrections above. You can reach us any time at hello@plainpatent.com.