A working SEO data library
The data I use when building sites for global markets: websites that are growing, keyword roots, backlink sources, and write-ups of what happened when I acted on them. Updated regularly, with the source and method stated for every dataset.
This month
Random preview · 10 free sites per issue. Pro unlocks all sites and the original ranking.
See all 186 sitesLast updated 2026-10-04
Workflow
Field notes
How to install Claude Code, sign in, choose a permission mode, write a CLAUDE.md, and roll back when something goes wrong. Based on the official docs as of October 2026.
Where to download the ChatGPT desktop app and run your first task, how to install and sign in to the CLI, what auth.json is, how sandbox and approvals work, and how to write AGENTS.md. Based on the official docs as of October 2026.
Toolbox
The toolbox is reorganized into eight steps that follow how a site gets built (find demand, analyze data, APIs, build, measure, monetize, promote, learn), growing from 37 to 94 tools, each marked with cost and whether I use it daily.
Rankings
The three rankings now go beyond 50 sites each: 186 in DR × Traffic Dual-Growth, 99 in New-Site Traffic Growth, and 200 in Paying New-Site Traffic.
Backlinks
Where to Submit list published: 3,088 channels merged and de-duplicated from several lists, filterable by type, audience, cost, DR, and observed link attribute, with markers for channels competitors use and ones I have submitted to.
Keywords
Keyword roots page published: 51 classic roots, 19 action roots with data, 16 candidate roots still to be exported, and 21 recent signal keywords, from the September 2026 snapshot.
Ship One, Ship Fast
Build one thing, ship it fast, and publish the reasoning, every action taken, and the Search Console curve together.
