Innovation Metrics Guide: Reading PlainPatent Scores

PlainPatent summarizes a historical record of US utility patents granted from 2015 through 2025. It does not measure applications, pending patents, current research activity, commercial success, or legal strength.

The four displayed inputs

Portfolio volume is the number of recorded grants assigned to an organization in the dataset.

Grant velocity compares two grant-year windows: 2020–2025 and 2015–2019.

Technology breadth is the number of CPC subclasses assigned to those grants.

Average claim count is the average number of claims per recorded grant.

These inputs are normalized and weighted 40%, 20%, 25%, and 15%, respectively, to produce the displayed score.

What the score can show

The score makes the same four historical measurements easier to compare across assignee records.

A higher score means a record ranks higher under this published formula relative to other records in the dataset.

It can be useful as a starting point for locating a company page or a CPC classification to inspect.

It is not an assessment of whether a company is more innovative, better managed, more valuable, or more likely to succeed.

Read grant timing carefully

Patent grants are not application filings. A grant may arrive years after an application was submitted.

For that reason, a change in grant-year totals does not establish a current R&D budget, product launch, market move, or strategic change.

Use current company disclosures and primary sources when those questions matter.

Read claim counts carefully

A claim count is a descriptive document feature, not a measure of claim scope, validity, enforceability, licensing value, or design-around risk.

Claims need legal and technical interpretation in their complete context. An aggregate average cannot replace that work.

For legal questions, consult the individual patent and a qualified professional.

Source and coverage

PlainPatent uses USPTO PatentsView grant and Cooperative Patent Classification data for the covered period.

The source covers US utility grants in this dataset, not design, plant, foreign, pending, or unrecorded patents.

Assignee names and classifications follow the source data and can be revised by the upstream provider.

A repeatable reading checklist

Confirm that the page identifies grants rather than applications.

Confirm the years included in the displayed comparison.

Check whether the figure counts grants or CPC assignments.

Keep an assignee total separate from a classification-assignment total.

Read the formula before comparing two score values.

Treat the output as a sorting aid, not a recommendation.

Do not infer legal scope from a claim-count average.

Do not infer current work from a grant-year change.

Do not infer business value from portfolio volume.

Use the underlying record when an individual patent matters.

Use a current primary source when a company’s activity matters.

Record the source vintage when saving a comparison.

Recheck the methodology after a source refresh.

State the limits alongside any summary you publish.

Keep the grant date distinct from any filing date.

Use cautious language for every conclusion.

See the methodology for the current source scope and processing notes.