Read it, then pass the check at the end to open step 3.
Mid game: read your site from the API, not the dashboard
Updated
Do this step before you form a theory. On 2026-10-15 the Page indexing report for clevrstart.com said 82 pages were not indexed: excluded by noindex 12, page with redirect 4, redirect error 2, not found 1, crawled not indexed 29, discovered not indexed 34. The same afternoon the URL Inspection API, asked about each of the 105 URLs in the sitemap, said 25. The dashboard counts every URL Google ever found on the domain property, including subdomains and dead paths; the API counts your sitemap. The full sort of those 82 rows is in google search console not indexed, which shows which rows were app subdomains, redirects and old sitemap paths.
The read, as I run it: a script called search-read.ts, in my own operating system, authenticates with a service account you add as a user on the property, reads the sitemap, calls the inspection endpoint once per URL in sequence, and writes a CSV. The schema is the procedure’s input:
url, verdict, coverage, last_crawl, google_canonical, user_canonicalA run over 105 URLs takes about ten minutes. On 2026-10-14 one API 500 killed a run after ten minutes with nothing written; on 2026-10-15 I added a retry-once-then-skip, commit 5af09df7, so a single failure costs one row, not the run. Google’s URL Inspection API documentation sets the quota at 2,000 inspections per property per day, which the docs show is plenty for weekly reads of a site this size.
Two limits to know before you trust the performance side. The performance data lags about three days. And it omits the query for roughly half of clicks, so the per-page file is the truth and the per-query file is a sample.
If you do not want to write a script, the console will take a submission by hand: themeisle’s step-by-step guide shows how to submit a sitemap and inspect a URL in the console, and I agree that is where to start if you have ten pages. At a hundred, one call per URL is the only way to read the canonical column for every page, and the canonical column is what disproved my first theory on clevrstart in one query.
Check: your CSV has one row per sitemap URL, and the google_canonical column is filled for every indexed row. If google_canonical ever differs from user_canonical, that row is your first repair.
Check what you read in “Mid game: read your site from the API, not the dashboard”
Three questions drawn from a larger pool, so no two papers are quite the same. Two right passes it and opens the next step. The questions are public; scoring them needs an account, and the key never leaves the server.
Loading the check…