The blog · Implementation · 7 min
What can you use Clay for in B2B prospecting?
Six jobs Clay does in a prospecting system, from finding leads where they talk to routing replies, with the names Clay gives each feature today.
Clay is a data and research platform for go-to-market teams: it finds companies and people, enriches them from many providers in one table, watches them for changes, and lets an AI model research and write against every row. In B2B prospecting that covers six jobs, finding leads where they actually are, filling in contact data, catching buying signals, researching accounts, writing outreach and routing replies. We are a Clay Solutions Partner and use it in nearly every system we build; these are the uses we run most, with the names Clay gives them today.
What can Clay do for B2B prospecting, at a glance?
Clay does six jobs in a prospecting system, and most teams use one or two of them. The table maps each job to the Clay feature that does it, under its current name, checked against Clay's own documentation in October 2026.
| Job | Clay feature | What it gives you |
|---|---|---|
| Find leads where they talk | X, Google Maps, YouTube and Apify sources | People and companies from the places your buyers are active |
| Find more companies like your best ones | Company Lookalikes | A new target list modeled on accounts you already win |
| Fill in contact data | Waterfall enrichment | Emails and phone numbers tried across providers in sequence |
| Catch buying signals | Signals: job changes, promotions, new hires, news and fundraising | A dated event on an account worth acting on |
| Research an account | Claygents and Use AI | Answers from the company's site and the web, per row |
| Act on it | HTTP API, Slack, Webflow | Leads sent to a sequencer, replies routed to a rep, a page per account |
Most B2B teams actively touch only 15 to 20 percent of their addressable market, and around 60 percent of an SDR's time goes to research and list-building. [source: The Demand Compass, chapter 1] Every row in that table is time taken off that 60 percent.
How do you find leads outside LinkedIn with Clay?
Clay finds leads outside LinkedIn by searching the places your buyers already talk and turning the results into a table. The X integration finds profiles and posts by keyword, so you can list the people discussing a topic in your field. Google Maps pulls local businesses by category and location, the right source when you sell to clinics, agencies or shops rather than enterprises. The YouTube source lists channels and videos on a topic, useful for creator partnerships. For anything Clay does not cover natively, an Apify actor run can be imported straight into a table: we use that to bring in job posts, Reddit threads and directory listings.
The question to ask first is not which source is biggest but where your buyers are visible. A list built from people who just posted about your problem starts warmer than any database export.
How do you find more accounts like your best customers?
Company Lookalikes finds more accounts like your best customers by taking a company you already sell to and returning companies that resemble it. You can also search by a description of the business, for instance "go-to-market agencies serving US startups", and get the companies whose own description matches. The list is a starting point: every company on it still needs qualifying against your profile before anyone reaches out, because similar on paper is not the same as in market.
How does waterfall enrichment find more emails?
Waterfall enrichment finds more emails by asking several data providers in turn and stopping at the first verified answer. If the first provider has no email for a person, Clay tries the second, then the third, and the same works for phone numbers. No single provider covers a whole list, so a waterfall returns more valid contacts than any one tool, and you choose the order, putting the provider you trust most for your market first.
It also removes a manual loop many teams still run: export a CSV from one tool, upload it to another to find emails, upload the result to a third to verify them. In Clay that is one column.
Which buying signals can Clay monitor?
Clay's Signals monitor a list for the events that mean an account is getting ready to buy: job changes, promotions, new hires, and news and fundraising, plus custom signals you define. A champion moving to a new company, a new head of sales arriving, a funding round closing: each is a buying trigger, a dated reason to talk to someone now rather than whenever the sequence says. Clay can post each new signal to a Slack channel so the right rep sees it the day it happens.
Signals are only useful if they are filtered. In one of our deployments, 829 signals were captured in a quarter and 150 were surfaced to a person; 82 percent were filtered out on purpose. [source: automate rev.ops., The Demand Compass, chapter 18] More on choosing them in what buying triggers you can automate and how to prioritize buying triggers.
How do you use AI inside Clay for research and writing?
AI inside Clay runs one prompt per row, so the same research or writing task happens for every account at once. Claygents read a company's website and the open web to answer a question you set, for example what the company sells, who it sells to, and whether it has published anything about the problem you solve. Use AI connects GPT, Claude or Gemini to any column, including with your own API key.
Two uses pay back fastest for us. The first is meeting preparation: when a meeting is booked, Clay researches the person and the company and writes a short brief into the CRM, so the rep walks in prepared instead of spending fifteen minutes searching. The second is outreach copy built from the research: the opener references something true and recent about the account, not a guess. Cold email with no relevance replies at 1 to 3 percent in our sends, while an audience that had already engaged with us replied near 20 percent. [source: automate rev.ops., our own sends, The Demand Compass, chapter 18] Relevance is what moves the number, and research is where relevance comes from.
Relevance is what moves the number, and research is where relevance comes from.
How do you send Clay's output to the rest of the stack?
Clay sends its output to the rest of the stack through actions on each row. The HTTP API action posts a lead to a sequencing tool or a CRM; the Slack action posts a message, or asks a person to approve one before it goes. We use the same pattern for replies: a reply comes into Clay, a model reads whether it is positive, and the positive ones go to the rep in Slack with the account's context attached. The Webflow integration creates a collection item per row, which is how a personalized landing page per account gets built from table data.
What has changed in Clay since 2025?
Clay has added a lot since our original video in May 2025. Sculptor builds workflows from a description in plain language. Audiences keeps a living list of accounts that updates as signals arrive. Recipes were renamed Functions. Lookalikes now cover people as well as companies. The jobs above have not changed; the shortest way to them has.
Related questions
Is Clay a CRM?
No. Clay finds, enriches and researches records and sends them on; the CRM is where your team keeps the relationship. Most setups write Clay's results into the CRM through an integration or the HTTP API.
Do you need to code to use Clay?
No for most of the jobs above, which are point and click. Formulas and the HTTP API go further, and an AI model can write the formula from a description. Sculptor now builds a first workflow from plain language.
Is it legal to email people Clay finds?
That depends on the countries you sell into, not on the tool. Rules for business email differ between the United States, the EU and the UK, so check what applies to your market, and keep an unsubscribe in every message.
What should you build first in Clay?
Build the one table that removes the most manual research for your team this week, usually enrichment of an existing list or meeting preparation. Signals come next, once you know which events actually precede your deals.
