SEO Audit Checklist Pdf

I open Google Search Console and look at the Pages report first. If I see a sudden drop in indexed pages, that is where I spend my time. Then I run URL Inspection on five key pages – the homepage, a category page, a product page, a blog post, and a landing page. I check that each is indexed, that the canonical tag points correctly, and that the reported URL matches the one I expect. I have caught misconfigured [canonical tags](/canonical-tag/) on client sites where the tag pointed to a URL that later 301-redirected, creating a loop. Fix the canonical before anything else.

Next I verify [robots.txt](/robots-txt/). I had a case where a client's /products/ path was accidentally blocked by a leftover Disallow from a staging site. Google could not crawl the catalogue for two months. I now check every rule against the sitemap. The sitemap itself must list only 200-status URLs with the correct <lastmod> dates and no conflicting directives. I use the Sitemaps report in Search Console to spot discrepancies.

[Core Web Vitals](/core-web-vitals/) are non‑negotiables now. I run PageSpeed Insights on the same five pages. If LCP is above 2.5 seconds or CLS above 0.1, I log it as a priority fix. Mobile usability comes next: viewport meta tag, tap targets at least 48px apart, no intrusive interstitials. I test on a real device, not just the emulator, because emulators miss font‑size and scroll‑blocking issues.

Finally, I validate structured data with the Rich Results Test. A client's product markup had a missing price field, so the reviews snippet never appeared. I also check for duplicate or conflicting schema types – especially breadcrumbs and FAQ schemas – which I validate against Google's [schema markup](/schema-markup/) guidelines. I log everything in a spreadsheet with a severity score so I can prioritise the fixes that move the needle.

An SEO audit can expand to fill any amount of time. I stop when I have confirmed that the critical crawl path – from homepage to the deepest content page – is clear, and that the top 20 traffic pages pass the basic checks. Anything beyond that is diminishing returns for a single audit.

Counter‑argument: 'But what about the long tail of 50,000 product pages?' I sample them. I select 100 based on highest traffic, highest crawl depth, and most recent updates. If the sample shows no patterns, I trust it. If I find broken links or missing titles, I escalate to a full crawl. The key is to decide the sample size before you start, not after you are drowning in spreadsheets.

Decision rule: if the index coverage report shows zero errors and the top 10 pages have Core Web Vitals in the 'good' range, I call the audit complete. I also stop when I have documented three critical issues and five minor ones. More than that and the client – or I – get paralysed by the list. I used to keep auditing until I had found a hundred issues. That was a waste. Now I focus on the ones that matter for revenue and crawl efficiency, and I schedule the rest for the next cycle.

Edge case: a site that just underwent a major platform migration or a [301 redirect](/301-redirect/) overhaul. For those, I do not stop after the basic sample. I run a full crawl because the risk of broken paths is too high.

I ignored canonical tags on my first audit. They looked fine – every page had one. But I had not checked the target URL's final redirect chain. One client's canonical pointed to a URL that itself had a noindex tag. The page vanished from the index for three months. Now I always trace the canonical's destination to the final 200-status response.

I also skipped pagination checks until a client complained that category page 2 and page 3 had different content but no rel=prev/next. Google saw them as separate entities, diluting the authority. I now include pagination in every audit, checking for proper rel markers and that paginated pages are not blocked by noindex or robots.txt. [Duplicate content](/duplicate-content/) from pagination is still one of the most common issues I find on ecommerce sites.

I used to rely entirely on automated crawlers. They flagged missing meta descriptions as errors, but I missed the real problem: a single incorrect directive in robots.txt that blocked the entire /blog/ section. Manual checks catch what tools miss. I run a manual review of the first five pages in every section.

Finally, I did not schedule repeat audits. I treated the checklist as a one‑off. After a site migration, I ran a new audit and found 40 new issues that had been introduced. Now I audit monthly, even if I only have time for the top five checks. Consistency beats depth when you are starting out.