mirror of
https://github.com/superdesigndev/treg.git
synced 2026-10-02 03:24:35 +08:00
Jason Zhou
661a56f2eb
Wave 2 of marketing/pseo-ship-plan.md. Transcripts, video stats, channel stats, search, and comments: all `compare` form, all one platform, and the first cluster where one row is free (the official Data API on the reader's own connected Google account, $0.00 against a 10,000 unit daily quota). Titles were measured before they were written, on the free own-key Google Ads endpoint. `youtube transcript` 110,000/mo, `youtube channel stats` 1,300, `youtube transcript api` 880, `youtube comment scraper` 210. The `<thing> api` phrasings people reach for are near-dead: `youtube video stats api` 0, `youtube channel data api` 0, `youtube subscriber count api` 10. Volumes and the dead list are in the ship plan. `voices` comes from ~180 Reddit and X posts read by two subagents, with thirteen vendor clusters excluded. Every quote was checked verbatim against the raw dump and every URL confirmed to exist before it was published. Tests: the SERP-fit check now parametrises over every page that ships rather than a hand-kept list of six, so the remaining 54 pages are guarded as they land; a shape check refuses a page with an empty section; and a link check refuses a `related` label that is not on the menu.