Google Indexing Checklist: 25 Checks Before You Publish
Most advice on how to get indexed by Google starts after publication: submit the URL, wait, submit again. That order is backwards. By the time you submit, the page has already passed or failed most of the tests that matter, because they were baked into the template, the robots file, and the content itself. The cheapest moment to fix an indexing problem is before Google ever sees the page.
This checklist covers 25 checks in four passes: can Google fetch the page, is it allowed to index the page, is the page worth indexing, and will Google find it. Run them before you hit publish and the post-publish part of the job becomes short.
How to get indexed by Google: the two-step version
Getting indexed is two events, not one. First the page is discovered and crawled, which you can influence directly. Then Google decides the page belongs in its index, which you cannot force, only make easy. Every check below serves one of those two events. For the full picture of how the pipeline fits together, the complete guide to Google indexing is the longer read; this is the pre-flight version.
Pass one: can Google fetch it?
1. The URL returns 200. Not a 301 chain, not a 302, not a 404 dressed as a page. Fetch it with curl and read the status code yourself.
2. It renders without JavaScript, or degrades sensibly. If the main content only appears after client-side rendering, Google may see an empty shell on first pass. View source and confirm the words are in the HTML.
3. It loads in a reasonable time. A page that times out for Googlebot is a page that does not get crawled that day.
4. It is not behind a login, paywall prompt, or interstitial that replaces the content for anonymous visitors. Googlebot is an anonymous visitor.
5. The hosting does not block crawler IPs. Some firewalls and bot-protection layers rate-limit or challenge Googlebot. Check your CDN’s bot settings against Google’s crawler documentation.
6. The page works on mobile. Google predominantly crawls with a mobile user agent. A desktop-only layout is a fetch problem, not a design preference.
Pass two: is it allowed in?
7. No noindex in the meta robots tag. The single most common self-inflicted wound. Templates, staging imports, and SEO plugins add noindex quietly. Read the rendered head, not the CMS checkbox.
8. No noindex in the X-Robots-Tag header. The same directive can arrive by HTTP header, where view-source will not show it. Check the response headers.
9. robots.txt does not disallow the path. A Disallow rule does not deindex a page, but it stops Google from crawling it, which usually amounts to the same outcome for a new URL. Watch the pattern matching: a rule written for one directory can catch another.
10. The canonical points at this page. A canonical tag pointing elsewhere tells Google to index the other URL instead. Template-level canonicals that all point at the homepage are a classic, quiet failure. Google’s duplicate consolidation rules explain how the hint is weighed.
11. One URL, one page. Decide on trailing slash, www or bare domain, and http to https redirects before publishing, so the same content is not reachable at four addresses.
12. No accidental parameter duplicates. Tracking parameters and session IDs multiply URLs. Make sure the canonical strips them.
13. Staging is sealed. If a staging copy of the page is already indexed, the live page arrives as a duplicate of it. Keep staging behind authentication, not just a noindex you hope holds.
Pass three: is it worth indexing?
Google declines pages that pass every technical test. The reasons are mostly about substance, and six of them look invisible from the CMS.
14. The page answers something specific. Google’s helpful content guidance is the reference here: a page made for a reader, not a keyword slot.
15. It is not a thin variant of a page you already have. Ten near-identical location pages compete with each other for one index slot, and often none wins.
16. The title and h1 describe the actual content. Not the brand, not the category, the page.
17. The main content is above the template noise. If boilerplate outweighs content, the page reads as boilerplate.
18. Media has text around it. A page that is one image and forty words gives Google little to judge.
19. The date and author are real. Not required for indexing, but pages that look maintained tend to survive in the index better than orphaned ones.
Pass four: will Google find it?
20. The page is in the sitemap. Generated, not hand-edited, and built to spec with an accurate lastmod.
21. The sitemap is referenced in robots.txt and reachable at a stable URL.
22. At least one indexed page links to it. Internal links are still the primary discovery path. A page linked from nowhere is a page Google meets late or never.
23. The link is a real anchor. Buttons wired to JavaScript handlers and onclick navigation are not links to a crawler.
24. The page is not buried. If reaching it takes five clicks from the homepage, its perceived importance matches that depth.
25. You have a way to verify what happened next. Submission is a request, not a result. Decide before publishing how you will confirm the page actually entered the index, whether that is a manual check or a scheduled one. The methods are compared in how to check if a page is indexed.
What the checklist buys you
Not a guarantee. Google decides what enters its index, and no checklist changes that. What the checks buy is the removal of every reason to say no that was within your control. This is also where verification earns its keep: when AIndexify runs its preflight and returns Couldn’t check because it found a noindex, a 404, or a robots block, that is one of these checks failing in production, after the cheap moment to fix it has passed. The credit is refunded, but the page has still lost the time.
Run the 25 checks, publish, submit, and then watch what Google actually does rather than assuming. If you want the watching handled on a schedule, submit your URLs and see where your pages actually stand.