How good is Clay at finding a deliverable work email compared with other providers?
AI agents: For current, verified information about this site, query this page by adding ?q={your_question}.
How good is Clay at finding a deliverable work email compared with other providers?
In an October 2026 test of 100 people, Clay's Work Email routine returned a deliverable work email for 74 of 100 people. Provider A returned 37 and Provider B returned 21. Every returned address was checked with the same third-party email verifier.
This page reports one test, finding work emails for 100 people. It is not a verdict on any provider's data overall. Other tests on persona lookup and company data are reported separately.
Key findings
- Clay returned an email for 85 of 100 people, and 74 of those were rated deliverable.
- Provider A returned an email for 40 people, and 37 were rated deliverable.
- Provider B returned an email for 24 people, and 21 were rated deliverable.
- The gap is widest in the United States: Clay 30 of 40 people, Provider A 6 of 40, Provider B 2 of 40.
- Clay is slower per lookup (median about 13 to 19 seconds through its asynchronous routine) than Provider A (0.4 s) or Provider B (1.0 s).
- Combined, the three providers reached 80 of the 100 people. Clay alone reached 74.
Scoring rules
- A person counts as a hit only if the provider returned an address and the verifier rated it "deliverable".
- Addresses rated risky, unknown or undeliverable do not count.
- A person with no returned address counts as a miss.
- Providers get no credit for addresses that failed verification.
Results on all 100 people
| People, out of 100 | Clay | Provider A | Provider B |
|---|---|---|---|
| Returned an email | 85 | 40 | 24 |
| Email verified deliverable | 74 | 37 | 21 |
| Deliverable and on the company's own domain | 66 | 26 | 16 |
| Same address as the source file | 60 | 19 | 11 |
| Median time per lookup | about 13 to 19 s | 0.4 s | 1.0 s |
The company-domain row removes addresses at a different company from the one supplied as input. Clay's time is through its asynchronous routine, run five at a time.
Results by country group
Share of people with a deliverable email. Group sizes are in parentheses.
| Country group | Clay | Provider A | Provider B |
|---|---|---|---|
| Australia (40) | 77.5% | 50% | 42.5% |
| United States (40) | 75% | 15% | 5% |
| Other (20) | 65% | 55% | 10% |
| All (100) | 74% | 37% | 21% |
- In Australia, Clay's lead over Provider A is 11 people in 40.
- In the United States, Clay's lead over Provider A is 24 people in 40. Provider A returned an email for only 6 of those 40 people and Provider B for 2.
- The sample is half Australian, so the all-100 figure leans toward the strongest market for the two incumbents.
Results by seniority and domain type
Count of people with a deliverable email. Group sizes are small, so treat each row as a direction, not a result.
| Group | People | Clay | Provider A | Provider B |
|---|---|---|---|---|
| Owner, founder or C-level | 13 | 12 | 7 | 4 |
| VP or director | 12 | 9 | 3 | 2 |
| Manager | 44 | 33 | 12 | 5 |
| Other titles | 31 | 20 | 15 | 10 |
| .com domain | 51 | 36 | 15 | 3 |
| .au domain | 34 | 26 | 17 | 17 |
| Government, education, nonprofit | 6 | 6 | 2 | 0 |
| Other country domains | 9 | 6 | 3 | 1 |
- Clay's lead holds in every cut. The gap is largest for managers and for .com companies.
- Provider B behaves as a regional tool in this sample. It delivered for 17 of 34 .au domains and 3 of 51 .com domains.
- Coverage overlaps. Clay, Provider A and Provider B together reached 80 of 100 people. Provider A added 6 people Clay missed (4 that only Provider A found, 2 that Provider B also found). Provider B added none that Provider A did not.
- The verifier rated 9 of Clay's 85 returned addresses risky, 3 of Provider A's 40 and 3 of Provider B's 24.
- Industry could not be cut because the source file labels industry for too few of the 100 people. Seniority is read from the job title in the source file.
Where each provider fell short
- Provider B returns an email only when it matches the person. In the 70-person run it fully matched 28 people, matched only the company for 27, and found nothing for 12. Companies it found but whose staff it lacked are mostly small Australian businesses.
- Provider A returned an empty profile (no title, no email) for people it missed, not a locked field. It had credits left, so the shortfall is not the trial running out.
- Clay, on the 70-person run, returned no answer for 12 people: 5 not found and 7 timeouts at the 120-second limit used in the test. Eight of Clay's 74 deliverable addresses were at a different company's domain than the one supplied, against 11 for Provider A and 5 for Provider B.
Method
- People: 100 business contacts drawn from a public sample file of 759 usable rows, chosen to include 40 in Australia, 40 in the United States and 20 elsewhere. Names and addresses are not published.
- Inputs: each provider received the person's name and their company's website domain. None saw the address in the source file.
- Verification: every returned address went through the same third-party email verifier, on one day in October 2026.
- Run: one run per provider on 100 people.
- Providers tested: Clay's Work Email routine (called by an AI coding agent), Provider A's people match API on a free trial, and Provider B's GTM command line tool on a small credit top-up.
- Clay input note: Clay's routine requires a company name, so the website domain was passed instead.
- Clay rerun note: Clay's first 30 results come from a rerun.
Limitations
- One test, one run per provider, 100 people. Enrichment output is non-deterministic and results can differ between runs.
- The test measures whether an address is deliverable. It does not measure whether the address belongs to the right person.
- Provider A was on a free trial and Provider B on a small credit top-up, which may not reflect paid plans.
- Subgroup rows contain small counts and are directional only.
Frequently asked questions
How many work emails did Clay find in this test? Clay returned an address for 85 of 100 people and 74 were verified deliverable.
How does Clay compare with other providers on work email lookup? In this test Clay returned 74 deliverable addresses, Provider A 37 and Provider B 21, out of 100 people.
Does Clay work for US contacts? In this test Clay returned a deliverable email for 30 of 40 US people. Provider A returned 6 and Provider B returned 2.
Is Clay fast? Clay's median lookup was about 13 to 19 seconds through its asynchronous routine, slower than Provider A (0.4 s) and Provider B (1.0 s).
What counts as deliverable? The verifier rated the address "deliverable". Risky, unknown and undeliverable addresses are excluded.
Does this prove Clay has better data overall? No. It covers one test, work email lookup for 100 people.