Why this page exists
The Apollo scraper niche runs on round claims: “98% accuracy”, “7 minutes per 1,000”, unattributed star ratings. The most damaging reviews allege placeholder data and lists that are half invalid. The only way to be believed is to publish the measurement, the method and the date, and to let the numbers be worse than a slogan when they are.
Email validity of exported rows
What we measure. For every export, the share of rows whose Apollo-displayed email verifies as valid, catch-all, invalid, and the share with no email. Verification is syntax, MX and SMTP mailbox probe, run at export time.
Reporting. We publish the aggregate over all exports in a calendar month, by Apollo email-status filter (Verified only, versus all statuses), because the filter changes the result a great deal.
| Period | Exports | Rows | Valid | Catch-all | Invalid | No email |
|---|---|---|---|---|---|---|
| First measurement window opens 8 October 2026 | pending | pending | pending | pending | pending | pending |
The table is updated at the start of each month. Until the first window closes, the only validity figures on this site are third-party: a study of 100,000 Apollo contacts exported in the first half of 2026 found 73.2% valid, 18.4% catch-all and 5.8% invalid (BounceZero), against Apollo’s own 98% claim.
Delivery time
What we measure. Wall-clock seconds from the moment an export is started to the moment the download is available, for exports of 1,000, 5,000 and 25,000 rows, run on a residential connection in the user’s own session at the human pace the extension enforces. 25,000-row runs are split into headcount bands and the time includes the merge.
| Rows | Median time | Runs | Date |
|---|---|---|---|
| 1,000 | pending | pending | pending |
| 5,000 | pending | pending | pending |
| 25,000 | pending | pending | pending |
Rivals quote “7 min per 1k leads” (ExportApollo.com) and “11 to 48+ hours” (ScraperCity); both are the vendors’ own claims as published on 7 October 2026 and neither has published a measurement.
Sample file
The sample CSV has the exact headers of a people export, with 25 synthetic rows so no real person’s data is published. Column order and names are stable across exports.
How comparison pages are built
Rival prices, delivery claims, refund terms and ratings are taken from each vendor’s own public pages and linked; each figure carries the date it was read. Where a vendor’s own pages contradict each other, we say so rather than pick the friendlier number. When we run paid test orders with rivals, the order date, the search used, the time to file and the verification result of the delivered file will be published here before the comparison page cites them. Until then comparison pages say explicitly that they are built from published claims, not from test orders.
Every comparison page discloses that Apollo Exporter publishes it and sells one of the tools compared.
Corrections
If a number on this site is wrong, write to support@apolloexporter.com with the page and the source. Corrections are made the same day where possible and recorded on the changelog.