Find emails from a website list in Clay with an HTTP API column

By Keencrest · 2026-11-01

You have a Clay table with a Website or Domain column and want a contact email for each row without spending Clay data credits on a provider lookup. Clay can call any HTTP endpoint from a column, and Apify exposes one that runs an Actor and returns the results in the same response, so each row makes one call and gets its answer back.

Use your own token. Create an API token in Apify under Settings, Integrations, and put it in the tool's secret or credential store. Never paste it into a shared sheet or a public workflow. The examples say YOUR_APIFY_TOKEN; replace it. Sending the token in the Authorization: Bearer header is safer than putting it in the URL, because URLs end up in logs.

What we tested: the HTTP request shown, run against the live Actor with a spend cap. One site per call, as Clay makes it. The click-by-click steps in Clay follow its public documentation and were not run inside Clay itself, so menu names may have moved; the request is what matters.

The request

Clay's HTTP API column takes a method, a URL, headers and a body, and lets you insert values from other columns. Configure it as:

Field	Value
Method	POST
Endpoint	`https://api.apify.com/v2/acts/datagleaner~website-contact-details-scraper/run-sync-get-dataset-items?maxTotalChargeUsd=0.05&fields=domain,bestEmail,status`
Header	`Authorization` = `Bearer YOUR_APIFY_TOKEN`
Header	`Content-Type` = `application/json`
Body	`{"websites": ["<Website column>"], "maxPagesPerSite": 4}`

Insert the Website column inside the quotes in the body. The same call as a command line, which is what we ran:

curl -X POST "https://api.apify.com/v2/acts/datagleaner~website-contact-details-scraper/run-sync-get-dataset-items?maxTotalChargeUsd=1&fields=domain,bestEmail,status" \
  -H "Authorization: Bearer YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"websites":["mozilla.org"],"maxPagesPerSite":4}'

Response from that call, 3.6 seconds end to end:

[{"domain": "mozilla.org", "bestEmail": "trademark-permissions@mozilla.com", "status": "ok"}]

Getting the email into a column

The response is an array with one object, so the email lives at the first element's bestEmail. In Clay, run the column on a few rows, open the response, and use the option to extract a field into its own column, choosing bestEmail. If a row has no email the field is null and the column stays empty, which is the signal to try another route for that company.

Keep it cheap and safe

Keep maxTotalChargeUsd at a few cents per row: one site costs at most $0.002, so a cap of 0.05 guards against a mistake without ever biting.
Run on 5 rows before running the whole table, and turn off auto-update so edits do not re-run paid calls.
Clay has its own timeouts and rate limits; a single-site call finished in under 4 seconds in our test, which is well within what an HTTP column normally allows.

What comes back

A JSON array with one object per site. Output order is not guaranteed to match input order (in our test, sitemaps.org came back before mozilla.org), so always match rows on domain, never on position. A real response for three sites, with fields=domain,bestEmail,status,failureReason:

[
  {"domain": "basecamp.com",  "bestEmail": "jason@basecamp.com", "status": "ok", "failureReason": null},
  {"domain": "sitemaps.org",  "bestEmail": null, "status": "noContacts", "failureReason": "no_contact_found"},
  {"domain": "mozilla.org",   "bestEmail": "trademark-permissions@mozilla.com", "status": "ok", "failureReason": null}
]

Without fields each object is large (around 60 keys: every email with the page it was found on and a kind, phones, 14 social networks, contact-form URLs, pages crawled). Ask only for what you need. A site that publishes no email still returns a row, with bestEmail: null. Junk input such as "not a site" returns a row with status: "invalidUrl" instead of failing the call.

What it costs

The Actor charges $0.002 for each site where an email is found and nothing for sites without one. 1,000 sites with 600 hits is about $1.20, and a 4-site test is under a cent. Apify's free plan includes monthly credit. The maxTotalChargeUsd query parameter in the examples is a hard cap on what one call can spend: the run stops when it is reached, so a typo cannot cost more than you set. Expect an email on roughly half to two-thirds of small-business sites and less on large brands; in our five-site test three sites gave one.

Try it: Website Contact Details Scraper on Apify. $0.002 per site with an email, nothing for sites without. The Apify free plan includes monthly credit, so a test run costs nothing out of pocket.

Related guides

All Keencrest guides