# Changelog of SimilarWeb Fast Scraper — Website Traffic, Rankings & Analytics (`pro100chok/similarweb-scraper`) Actor

- **URL**: https://apify.com/pro100chok/similarweb-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/pro100chok/similarweb-scraper.md

## Changelog

### 0.0 (2026-09-10)

- **Up to 1,000 domains per run** (was 50). Concurrency was re-tuned from load tests on 1,000+ domains, so a full run finishes in minutes instead of hitting the timeout.
- Fixed a data-loss bug: requests failing with a bad proxy TLS certificate (or an HTTP/2 stream error) were not retried and the domain was dropped silently. They now retry on a fresh connection, which lifts delivery to ~100%.
- Keyword density now returns up to 5,000 phrases per phrase-length group, ranked by density. This keeps items under the platform's 9 MB per-item limit on very large pages.
- Run log now reports progress and a final delivered/expected item count.

### 0.0 (2026-08-30)

- Paying Apify users are now routed through a premium rotating residential proxy pool — higher success rate and noticeably fewer 429s on large runs. Free users keep running on Apify Residential.
- Unified codebase: both SimilarWeb Actors now run the exact same fetching and parsing logic.

### 0.0 (2026-07-07)

- Initial release. Batch-analyze 10-50 domains per run with three search types selected via `searchType`:
  - `similarweb` - traffic and engagement (GlobalRank, CountryRank, CategoryRank, Engagments, EstimatedMonthlyVisits, TrafficSources, TopCountryShares).
  - `similar_sites` - competitor / alternative sites (SimilarSites with visits and category rank).
  - `aitdk` - WHOIS records plus 1-5 word keyword density.
- Added dataset schema with field definitions and Overview / Similar Sites / AITDK output views.