Crunchbase API review

Four ways in, and picking the wrong one is the usual mistake

Crunchbase does not meter per call, so there is no cost table to build here. What there is instead is a choice of endpoint family, and getting it wrong is what turns a tidy integration into a rate-limit problem. The API is read only throughout: there is no write side, so this is a source and never a destination.

The order that works is autocomplete once to resolve your account list to permalinks, store those permalinks, then entity lookups from then on. Teams that skip the first step end up searching by name on every run, which costs several calls per company instead of one and burns the 200-a-minute ceiling on matching rather than on data.

Crunchbase API pricing, plan by plan

There is nothing to tabulate here, and that is the finding rather than a gap in the research. Crunchbase publishes no API pricing of any kind, so the Crunchbase API cost depends entirely on what you negotiate.

This directory's rule is that a number has to come from the vendor's own page or it does not get printed, and that rule bites hardest here. Treat any Crunchbase API price you find in a blog post as somebody else's negotiated rate rather than a list price, because there is no list.

One number, and it applies everywhere

Crunchbase API rate limits are the one operational figure Crunchbase does publish: 200 calls a minute across its API endpoints, which it describes as protecting platform reliability rather than as a tier of service.

Two hundred a minute is generous for a signal workflow, where you are looking up the handful of companies that did something this week, and tight for a backfill, where you are looking up all of them. Size the first run against the ceiling rather than against the deadline: 100,000 entity lookups is about eight hours whatever else you do, so it wants starting on a Friday rather than discovering the arithmetic on a Monday.

The attribution rule is the one that catches product teams by surprise, because it reaches outside the integration. A dofollow link, visible to crawlers, sitting next to the data it credits, is a design and an SEO decision rather than an engineering one, and it has to be agreed before anybody renders a Crunchbase field in a customer-facing screen.

How to create a Crunchbase API key

The Crunchbase API key is called a user key and travels in one of two ways. Crunchbase describes the API as being for approved developers, which is consistent with access starting at a sales conversation rather than a signup form.

The approved-developer framing is the part to plan around rather than the choice of header. There is no self-serve key, so the lead time on this integration is a sales cycle rather than an afternoon, and that belongs in the project plan next to the engineering estimate.

What a Crunchbase API call looks like

The base URL is api.crunchbase.com/api/v4 and the key travels one of two ways: as a user_key query parameter or as an X-cb-user-key header. Prefer the header. A key in a query string ends up in access logs, proxy caches and anything else that records a URL, and Crunchbase documents both forms without ranking them.

# 1. one organization, by its permalink
GET https://api.crunchbase.com/api/v4/entities/organizations/crunchbase
X-cb-user-key: $CB_USER_KEY

# 2. the same call with the key in the URL, which Crunchbase
#    documents too. It works, and it leaks into every log.
GET /api/v4/entities/organizations/crunchbase?user_key=...

# plain HTTP is refused rather than redirected:
# 426 Upgrade Required

Where the Crunchbase API documentation lives

Crunchbase runs two documentation sites at once, a current one and a v4-legacy one, and search engines will happily send you to the wrong one. The Crunchbase API documentation worth bookmarking is the first row below.

The autocomplete endpoint is the underrated one for go-to-market work. Matching a company name as your CRM spells it to the right Crunchbase entity is the step every funding-signal pipeline needs and the step most of them get wrong, and there is a purpose-built endpoint for it.

Crunchbase also ships an MCP server, and it is on the MCP page with the others in this directory.

Crunchbase on its own API, in 29 seconds

Crunchbase publishes this on its own channel, so it is Crunchbase's account of Crunchbase, not ours. It is here because a rate card cannot show you the thing moving, and a minute or two of the product is faster than any paragraph on this page at telling you whether the shape of it fits yours.

Twenty-nine seconds, and an announcement rather than a walkthrough: Predictions and Insights, reachable through the API. The only substantial API video on the channel is a 2020 webinar on migrating to v4, which is the version this page covers and six years old.

Published 11 October 2024 on Crunchbase, the channel Crunchbase's own pages link to, and read on 13 October 2026. Nothing in it is a source for a price or a limit on this page: every figure in the tables above comes from a Crunchbase page.

Strengths and limits

Crunchbase is a dataset with a long memory, and that is what you are buying.

The pattern underneath those two columns is consistent: Crunchbase is strong on what happened to a company and weak on the things you can buy by the credit elsewhere. Buy it for the event history and fill the firmographics from a provider with a published rate, rather than paying a negotiated price for both halves.

If Crunchbase is not the answer

Four providers in the Signal APIs listing that solve the same problem differently. Each one is a trade, and the trade is named.

The full set, organized by the reason people leave, is on the Crunchbase alternatives page.

Before you build against it

Four things to settle first, each one read off Crunchbase's own pages. The ticks are remembered in this browser, so you can come back to a half-finished integration.

Who should actually buy this

Buy Crunchbase when funding is your trigger. A company that raised a Series B last month is a different prospect from the same company six months earlier, and this is the cleanest source for that event in this directory. If your pitch lands differently after a raise, the data is worth a conversation.

Go in knowing you cannot test first. The withdrawn free tier means your evaluation has to happen inside the sales process, so make it a condition: ask for a trial key against your own account list and count the matches before you sign. Anyone selling a dataset should be willing to let you measure it.

For everything that is not funding, look elsewhere first. Firmographics and headcount are available from several cheaper providers here with published rates and free tiers, and paying a quoted enterprise price for data you can buy by the credit is the usual way this purchase goes wrong.

If you do buy it, wire it as a trigger rather than as a lookup. The expensive mistake with a quoted dataset is putting it in the hot path of every enrichment, where the per-minute ceiling and the per-contract cost both bite at once. The cheap pattern is a scheduled job that watches your own account list for funding events and writes a flag, with the rest of the enrichment running on providers that charge by the credit.