A literally exhaustive list is impossible — private use is unobservable and “scraping in some way” has no finite public universe. This is a best-effort census of publicly documented organizations from a broad live-web scan completed 2026-09-18: 208 unique company or organization rows backed by 92 independent source hosts. Categories overlap, so the six category counts must not be summed as unique organizations.
One organization can appear in several categories (e.g. Exa and SerpApi), so bars overlap and do not add up to 208. “Other scraping/crawling/browser automation” is broad by definition, which is why it is the largest.
| Organization | Categories | Documented use (first of several, where multiple) | Sources | Source |
|---|
Method: best-effort census of publicly documented use, one row per canonical organization, compiled from a broad live-web scan completed 2026-09-18 across vendor customer pages, case studies, documentation, news, GitHub, and LinkedIn (1,459 + 1,339 + 1,646 candidate pages across three searches). Rows kept only where an identifiable company, nonprofit, university, or government body has an explicit public statement of actual use; mere integrations, tutorials, unsupported logo mentions, and vendors described only as sellers were excluded. 208 rows, 92 independent source hosts; the largest single URL backs 13 rows (6.3%). Counts measure publicly documented organizations — a lower bound, not every company in existence. The table shows the first documented use per row for space; category filters and search cover all rows.