.avif)
Build agents for any GTM task
Run GTM agents that reason over your data, and see exactly why they do what they do.
Agents built on your entire GTM context
Research accounts, iterate on prompts in real time, and monitor your book of business — with full visibility into every agent's decision.
Find unique data no provider has
Point Claygent at any manual GTM task like company research, lead qualification, signal detection. Get back structured data for custom plays, whether it’s niche company attributes, buying triggers, org changes, or contact info.
Test and iterate on prompts
Describe what you want in natural language, test against real data, and roll back instantly if needed. Changes you make to the prompt update everywhere that the agent runs.
Glass box, not black box
Every agent decision in Clay comes with a full reasoning trace. For agents run over Audiences, get a full view of your spend and errors per agent and per segment across every account in your book.
Up-to-date account intelligence for your book of business
Synthesize new information from your CRM, data warehouse, Gong calls, and other signals via Account Research Agents, available in Audiences.
Each agent comes pre-loaded with GTM context and remembers what it's learned, so you can target the right accounts and improve outbound plays.
How to create your first agent in Clay
Create custom agents by describing your task, connecting tools, and deploying to any table.
What our customers say about us
"Clay auto-enriches contacts with accurate addresses for our gifting program, saving half the time of our growth team."
"Clay-built lists helped identify over 55% of revenue sold with Agent Analytics (our new product) in its first quarter, driving two-thirds of its pipeline"
"This market research analysis would have taken my team 3 months without Clay. It was enlightening to see how speedy you could get to a V1 answer."
FAQs about agents in Clay
What is a Claygent? How are Account Research Agents different?
Claygents are agents added to a Clay table that take inputs, follow your instructions on a task, and write a structured output into your desired column.
Unlike Claygents in a table, Account Research Agents are run over Audiences in Clay and have the added benefit of:
- Unlocking high-volume plays: No cell size or row limits in Audiences. Context, scheduling, and enrichment come pre-configured, so you can run plays like outbound expansion or closed-lost analysis without rebuilding a pipeline for every new question.
- Improving with every run: Claygents in tables start fresh every run. Account Research Agents’ context lives in Audiences, so intelligence builds account by account instead of resetting each pull.
- Governing your agents. Audiences gives you one place to see and control your agent activity, including viewing agent runs, errors, and spend over time)
Here’s some quick rules of thumb of when to use Claygent in a table vs. an Account Research Agent:
What are the main use cases for Claygent?
Agents in Clay can be used flexibly for many GTM use cases - ones we frequently see from customers building in tables include:
- Account research & meeting prep so you go into calls with the right context on your prospects/customers
- Discovery & finding bespoke data in the form of web research, niche signals, proprietary data points
- Personalized outbound to run campaigns at scale
- Scoring & qualification on large prospect lists based on criteria unique to your business
For plays that you need to run at scale, try using an Account Research Agent over Audiences instead.
When do I use Claygent vs. “Use AI” in a table?
“Use AI” is for offline tasks like content generation. This could include tasks like copywriting, tone-matching, and categorizing existing data.
Meanwhile, Claygent has access to a range of tools that allow you to do more “last-mile” research. For example, web access allows your Claygent to do more real-time research. This could include tasks like analyzing competitor homepages or extract deeper firmographic context.
How much do agents in Clay cost?
Claygents in tables are available in all plan tiers. Account Research Agents (in beta) are available for Enterprise, Growth, and Launch plans only.
For all agents in Clay, customers are charged on the actual complexity and token usage of their requests. There are two components to the pricing:
- Data credits: Variable pricing for advanced reasoning models; fixed pricing for Clay’s own models and standard content generation. (See our model pricing reference)
- Action: 1 action per AI prompt
For Account Research Agents, data credits are variable pricing only and 1 action is charged per record processed by the agent. Exports to the CRM or data warehouse have no additional charge.