Patent data is one of those datasets that sounds niche until you realize how many people actually need it. IP attorneys doing prior art searches, competitive intelligence teams tracking what competitors are filing, researchers mapping innovation trends, startups checking freedom-to-operate before they build anything — they all need USPTO patent data, and getting it programmatically has historically been a pain.
The PatentsView API from the USPTO is a solid public resource, but it has its own query syntax and pagination quirks that take time to figure out. If you just want the data without wrestling with the API directly, there is a cleaner path.
What the Data Looks Like
PatentsView covers granted US patents going back to 1976. Each record includes:
- Patent number, title, and abstract
- Filing and grant dates
- Inventor names and locations
- Assignee (company) information
- CPC and USPC classification codes
- Citation counts
That is enough to do serious work — competitive landscaping, inventor network analysis, technology trend mapping, or just verifying whether a specific invention is already patented.
Who Actually Uses This
The use cases I see most often:
Competitive intelligence — track what patents a company or research group has been filing in a specific technology area. File dates tell you when they started working on something, not just when they finished.
Prior art research — search by keyword or classification code to find existing patents before filing or building.
Investor/M&A due diligence — understand a companys patent portfolio depth before making a bet.
Academic research — map innovation clusters, citation networks, or geography of invention.
How to Access It via Apify
Rather than building your own PatentsView client, you can use the USPTO Patents PatentsView Search actor on Apify. You pass in a search query, set a maxResults limit, and get back structured JSON you can drop straight into a spreadsheet, database, or pipeline.
The actor handles pagination, rate limits, and the PatentsView query syntax for you. You can run it on a schedule if you want to monitor new filings in a specific area over time.
{
"query": "machine learning semiconductor",
"maxResults": 50
}
Output fields include patent number, title, abstract, assignee, inventors, grant date, and CPC codes.
Getting Started
If you are already on Apify, just add the actor and run it with your search terms. If you are new to Apify, there is a free tier that covers enough runs to evaluate whether the data fits your workflow.
Patent data moves slowly compared to most datasets, but the signal-to-noise ratio is high when you are doing IP research. Worth having in your data toolkit.
Ava Torres is a data engineer based in SF building automation tools for lead gen and research workflows.
Top comments (0)