Guides
Google Search Console Audit: What to Read, in What Order
Search Console has a dozen reports and three of them answer most questions. Reading them in the wrong order is how people spend an afternoon and learn nothing.
Run a fresh DomainLens audit and use the report as your priority list.
The three reports that matter
Start with Pages every time. A performance report is meaningless for URLs Google never indexed, and there is no point optimising Core Web Vitals on pages that are excluded.
| Report | Answers | Read it when |
|---|---|---|
| Pages (index coverage) | What is indexed and what is excluded, with reasons | Always first |
| Performance | Which queries and pages produce clicks | After coverage is clean |
| Core Web Vitals | Which URL groups fail the thresholds | Once content is working |
| URL Inspection | Everything about one specific URL | When investigating |
| Sitemaps | Whether submission worked | After publishing a new sitemap |
Reading the coverage report properly
The report groups URLs by why they are not indexed, and the groups mean very different things. Some are correct and require nothing; others are bugs; one is a judgement Google made about quality.
The mistake is treating everything under "Not indexed" as a problem. A noindexed admin page appearing there is the system working.
- Excluded by noindex tag — intentional, unless it is not. Check a sample.
- Alternate page with proper canonical — correct; these are the duplicates you consolidated.
- Crawled, currently not indexed — Google saw it and declined. A quality signal, not a bug.
- Discovered, currently not indexed — Google has not even fetched it. Usually discovery or budget.
- Page with redirect — correct if you redirected it deliberately.
- Soft 404 — a real bug: your server says 200 and the page says nothing is there.
- Server error, blocked by robots.txt — real bugs, fix immediately.
The two states people misread
"Crawled — currently not indexed" is not a technical fault. Google fetched the page, evaluated it and chose not to index it. Re-requesting indexing does nothing; the page needs to be more useful, more distinct, or better linked.
"Discovered — currently not indexed" is different: Google knows the URL exists and has not fetched it. That points at crawl priority, internal linking or server responsiveness rather than content quality.
Turning the performance report into decisions
The position eight to fifteen band is where effort converts fastest, because the page already ranks and needs improvement rather than creation. See how to improve Google rankings for what to do with them.
- 1Filter to the last three months and sort by impressions with low clicks — those are pages ranking but not being chosen.
- 2Check the title and description of those pages; a poor snippet is the usual cause.
- 3Find queries at positions eight to fifteen — the cheapest movement available anywhere in SEO.
- 4Compare page-level clicks against your intent: a docs page taking commercial queries means the marketing page is losing.
- 5Ignore query-level totals reconciling against site totals; anonymised queries are excluded from rows and included in aggregates.
What Search Console will not tell you
It covers only verified properties, holds sixteen months of data, omits low-volume queries from row-level reports, and gives no explanation beyond a coverage state string. It also shows a limited backlink sample rather than a complete picture. For competitor and market data you need third-party tools, and for what a URL returns right now you need to check it directly. See Search Console vs third-party tools and URL Inspection .
How DomainLens contributes
Search Console tells you what Google concluded; it rarely tells you why. DomainLens supplies the page-side facts — status code, robots directives, canonical, rendering — so a coverage state has something concrete to be explained by. See indexability debugging .
- What does "Crawled — currently not indexed" mean?
- Google fetched the page and decided not to index it. It is a judgement about value, not a technical error, and requesting indexing will not change it.
- Why do query totals not match the site total?
- Queries below a privacy threshold are omitted from row-level data but still counted in aggregates. The gap is expected behaviour.
- How far back does Search Console data go?
- Sixteen months. Export regularly if you need longer history, because there is no way to recover it afterwards.
- Which report should I check first?
- Pages, always. Performance and Core Web Vitals are meaningless for URLs that are not indexed.