# Clay.com API Documentation

By [Florian Martens](https://pipe0.com/authors/florian-martens) · Published Aug 3, 2026

Source: https://pipe0.com/blog/clay-api-documentation

> Clay.com recently released their API, but the API documentation is not linked from their website. Find the clay.com API documentation.



In June 2026, [Clay](https://clay.com) released a public API. The documentation is
genuinely hard to find. It is not linked from their homepage, and it does not live with
the rest of the Clay docs.

Here it is:

**[developers.clay.com](https://developers.clay.com/)**

That is the answer to the question. The rest of this post is about what is actually in
there, because "Clay has an API" and "Clay's API does what you expect" are not the same.

## What is in the Clay API [#what-is-in-the-clay-api]

Clay's public API is small. Thirteen operations, three of them beta, one of them Enterprise
only.

They fall into three groups.

**Routines** run Clay logic: enrichment, research, scoring, or whatever else your team has
packaged up.

**Search** queries Clay's GTM database for people and companies.

**Tables** is an endpoint to control your Clay tables. This endpoint is Enterprise only.

## Why Clay's API is not a real enrichment API [#why-clays-api-is-not-a-real-enrichment-api]

In sales technology, technical terms are often used loosely. When it comes to APIs, it's
worth digging deeper to understand why the Clay API might not behave the way
you expect.

### Stateless enrichment APIs [#stateless-enrichment-apis]

A stateless enrichment API can be used to find data. You send an email address and the API responds with a phone number.

If you send the same email address again, you receive the same phone number. In the
background, your request does not create anything. Your request is forwarded to a database that either
has your data or does not.

Stateless APIs are the industry standard.

### Stateful enrichment APIs (Clay's approach) [#stateful-enrichment-apis-clays-approach]

Clay's API does not work this way. Instead, routines (Clay's enrichment functions) are things you create before you can send your data.

Enrichment only works if you've created and stored a procedure within Clay. After you have done so, you can invoke this procedure.

This is the same mental model as: Create Clay table > run Clay table.

### Is this bad design? [#is-this-bad-design]

No. But it's usually not what people ask for when they prompt "Does Clay have an API?".

Clay is built around tables, cells, and functions. Its API is a faithful representation of this model.

Instead of giving you or your agent the same enrichment capabilities as Clay, you are handed a remote control.

## Alternatives to Clay's API [#alternatives-to-clays-api]

At [pipe0](https://pipe0.com), we created a new data model to power Clay-like enrichment *and* offer a real API.
Our API works fully stateless while also being able to power Clay-like tables with millions of rows.

Connecting your agent to pipe0 will not give it a remote control but real enrichment capabilities.