Company Lookup - Crunchbase & Wikidata Company Profiles

Turn a search term into a table of companies: official name, what they do, their industry, headquarters, website, employee count and founding year. Search a niche such as "fintech" or "AI startups", or look up a specific company name.

No API key required.

What you get per company

Field	Example
name	Qonto
shortDescription	French business banking platform for SMEs and freelancers
categories	["financial services", "fintech"]
location	Paris, France
website	https://qonto.com/
employeeCount	1600
inceptionYear	2016
slug	stable identifier
crunchbaseUrl	only when a verified Crunchbase ID exists
wikidataId, wikidataUrl	the source record, for further enrichment
source	crunchbase or wikidata

Search filters

• searchQuery - the term to look up, e.g. fintech, AI startups, or a company name
• category - keep only companies whose industry matches, e.g. financial services
• location - keep only companies whose location matches, e.g. Berlin
• maxItems - how many companies to save

Example input

{
  "searchQuery": "fintech",
  "location": "Berlin",
  "maxItems": 50
}

Use cases

• B2B lead list building - get companies in a niche with their websites as the outreach entry point
• CRM enrichment - turn a column of company names into industries, locations and sizes
• Market mapping - collect everyone operating in a category and see how they position themselves
• Competitive research - find the peer set around a company through its industry tags
• Investor and scouting work - filter by founding year and size to isolate a stage
• Data cleaning - replace inconsistent company names with their canonical form

How the two sources work together

The actor tries Crunchbase first. Crunchbase blocks datacenter traffic, so that path only succeeds when you supply a residential proxy.

When Crunchbase is unavailable, it falls back to the Wikidata company graph, which is open, unmetered and reachable from anywhere. Wikidata is searched two ways, by company name and by full text, and the results are then filtered down to entities that are actually companies rather than concepts or people. Industry, headquarters and country references are resolved to readable labels.

This is why the actor returns useful results out of the box: the fallback is a real data source, not an error path.

Scope and honesty about limits

This is company lookup and profile enrichment. It does not export funding rounds, investor lists, valuations or cap tables. Those sit behind Crunchbase's paid API and licence, and no scraper should promise them.

crunchbaseUrl is only set when a verified Crunchbase identifier exists for the company. It is never guessed from a name, so a missing value means "no confirmed Crunchbase profile", not a broken link.

Notes

• Wikidata coverage is strong for funded, listed and notable companies and thin for very small local businesses. For local business data, a directory or map-based source is the better tool.
• employeeCount and inceptionYear are present when the source records them, which is common for larger companies and rarer for early-stage ones.
• The category and location filters are applied to the retrieved records, so a narrow combination over a small maxItems can legitimately return nothing.
• A run that finds no company fails with an explicit message rather than reporting an empty success.