Home › Guides › Source’s Brands in Chrome’s Traffic Ranks
✓ checked 16 October 2026 · sources: Chrome UX Report (BigQuery), npm downloads API, GitHub, GlobeNewswire · method · caveats · download the data
Honeygain’s user dashboard is the highest-ranked site among the brands in Source, the group that Oxylabs and four sister companies launched on 13 October 2026. In Google’s Chrome UX Report it was in the top 50,000 origins in September 2026, as it was a year earlier. The other 8 Source-brand sites with September data ranked between the top 500,000 and the top 1,000,000. Of the 9 with data in both September 2025 and September 2026, 1 moved up a bucket, 6 stayed in the same bucket and 2 moved down.
About Source. According to Source’s press release on GlobeNewswire, the group brings together Oxylabs, Decodo, Webshare, Honeygain and ScrapingBee, which “continue to operate independently”. The release says the ecosystem reached “a $3.6 billion valuation in a fundraise led by Warburg Pincus this year” and that AI products drive “a four-fold increase in demand across Source companies in the last year”. Those are the company’s figures; we haven’t verified them, and nothing on this page measures demand or revenue.
Honeygain dashboard, Sep 2026 top 50,000Bright Data MCP npm downloads, Sep 2026 38,804Source brand sites checked 10
Headline findings
- Honeygain’s two sites moved in different directions. dashboard.honeygain.com, where people who share their internet connection through the Honeygain app log in, stayed in the top 50,000 in all four months we checked. The marketing site, www.honeygain.com, went from the top 50,000 in September 2025 to the top 500,000 in September 2026.
- oxylabs.io moved down one bucket, from the top 500,000 to the top 1,000,000, between March and August 2026. Its dashboard stayed in the top 1,000,000.
- ScrapingBee’s main site dropped to the top 5,000,000 in August 2026 and was back in the top 1,000,000 in September. Decodo’s dashboard went the other way over the year, from the top 5,000,000 in September 2025 to the top 1,000,000.
- For comparison, brightdata.com, a competitor not in Source, was in the top 500,000 in every month we checked, the same bucket as www.webshare.io (top 500,000). Bright Data’s MCP server, which connects AI agents to its scraping tools, had 38,804 npm downloads in September 2026, 2.39 times its 16,203 in September 2025 and its highest month so far. The GitHub repository has 2,678 stars.
- source.ai itself isn’t in the Chrome UX Report yet. The latest release covers September 2026, before the launch.
Where each site ranked in September 2026
Rank buckets by month
| Brand | Origin | Sep 2025 | Mar 2026 | Aug 2026 | Sep 2026 |
|---|---|---|---|---|---|
| Oxylabs | oxylabs.io | top 500,000 | top 500,000 | top 1,000,000 | top 1,000,000 |
| Oxylabs | dashboard.oxylabs.io | top 1,000,000 | top 1,000,000 | top 1,000,000 | top 1,000,000 |
| Decodo | decodo.com | top 1,000,000 | top 1,000,000 | top 1,000,000 | top 1,000,000 |
| Decodo | dashboard.decodo.com | top 5,000,000 | top 1,000,000 | top 1,000,000 | top 1,000,000 |
| Webshare | www.webshare.io | top 500,000 | top 500,000 | top 500,000 | top 500,000 |
| Webshare | dashboard.webshare.io | top 500,000 | top 500,000 | top 500,000 | top 500,000 |
| Honeygain | www.honeygain.com | top 50,000 | top 100,000 | top 100,000 | top 500,000 |
| Honeygain | dashboard.honeygain.com | top 50,000 | top 50,000 | top 50,000 | top 50,000 |
| ScrapingBee | www.scrapingbee.com | top 1,000,000 | top 1,000,000 | top 5,000,000 | top 1,000,000 |
| ScrapingBee | app.scrapingbee.com | top 1,000,000 | – | – | – |
| Bright Data (not part of Source) | brightdata.com | top 500,000 | top 500,000 | top 500,000 | top 500,000 |
| Bright Data (not part of Source) | docs.brightdata.com | top 5,000,000 | top 5,000,000 | top 5,000,000 | top 5,000,000 |
For how Chrome UX Report ranks work and how AI chatbots moved through them, see AI chatbots in Chrome’s traffic ranks. For the countries Honeygain-style bandwidth-sharing apps draw on, see the home internet supply chain, and for the week of AI-scraping news this sits in, the scraping and AI week.
Method
On 16 October 2026 we queried the Chrome UX Report’s materialized.device_summary table in BigQuery for 16 origins (the brands’ main sites and dashboards, Decodo’s former Smartproxy domains, source.ai and two Bright Data origins) in September 2025, March 2026, August 2026 and September 2026, taking each origin’s rank bucket. The query processed 1.07 GB. Origins with no row weren’t in that month’s release, which happens when a site has too few Chrome visitors. Download counts are from the npm downloads API, by day, summed by month. Stars are from the GitHub API. A separate script recomputes every number from the saved responses.
Caveats
- Chrome UX Report ranks are coarse buckets (top 1,000, 5,000, 10,000, 50,000, 100,000, 500,000, 1 million, 5 million and so on), so one step can be a large or small change in visits.
- The data covers Chrome users who opted in to usage statistics, and counts page loads, not customers or revenue. A dashboard that customers visit every day can outrank a marketing site.
- Rank moves don’t say why traffic changed; site redesigns, domain moves and other sites’ growth all shift buckets.
- npm downloads include automated installs. The daily series shows zero downloads on 3 Sep, 7 Sep, 8 Sep, 15 Sep, 6 Oct, which look like gaps in npm’s data, so the September 2026 total may be an undercount; the last days before we checked weren’t reported yet.
- Source’s valuation and demand figures are from its press release and have not been independently verified.
Download the data
- Source brand rank buckets (CSV): one row per origin and month.
- Bright Data MCP npm downloads by month (CSV), April 2025 to 15 October 2026.
Cite as: ProxyPicker, “Source’s Brands in Chrome’s Traffic Ranks”, 16 October 2026, https://proxypicker.com/source-brands-chrome-ranks/. Data: Chrome UX Report (CC BY 4.0), npm, GitHub.
Related: AI chatbots in Chrome’s traffic ranks · Home internet supply chain · Scraping and AI week.
Press & data requests: email [email protected] for interviews, the raw data behind this page, or corrections. See also our contact page.