Technical Audit SEO

I've seen too many sites with excellent content that never rank because the technical foundation is rotten. A technical SEO audit is the systematic check of that foundation: crawlability, indexation, rendering, performance, and structure. It's not about keywords or backlinks — those come later. For me, an audit answers one question: can Google see the site the way I intend? I start with Google Search Console's Coverage report. If it shows 5,000 valid pages but 12,000 excluded, I know something is off. Then I run a crawler like Screaming Frog to inventory every page. I look for HTTP status codes, missing meta tags, duplicate content, and canonicalisation issues. One number I always note: the ratio of indexed pages to total pages. If that ratio is under 80%, something needs fixing. Some practitioners skip the rendering check, but with [JavaScript SEO](/javascript-seo/) heavy sites, that's a mistake. I've seen pages where Googlebot's rendered HTML was empty because a fetch request timed out. A good audit catches that. And here is a decision rule: if you have limited time, prioritise issues that block indexing over everything else. A page that isn't indexed can't rank, no matter how good the content.

A technical audit is most valuable when circumstances change. Launching a new site? Run an audit before you submit a [sitemap](/sitemap/). I once found that a client's new site had a chain of three 301 redirects on the homepage, adding 600ms of load time. Fixing that improved both user experience and crawl efficiency. After a redesign or migration, an audit becomes essential. I use a crawler to map old URLs to new ones and check for redirect loops. One audit uncovered that 40% of old URLs were returning 404s because the redirect rules were applied only to a subset of pages. Traffic drops are another trigger. If rankings tank, an audit often reveals the cause — perhaps a misconfigured [robots.txt](/robots-txt/) accidentally blocking key sections. I recall a case where a site lost 70% of organic traffic overnight. The culprit was an accidental 'Disallow: /' added during a CMS update. Performance also matters. Since [Core Web Vitals](/core-web-vitals/) became a ranking signal, I always include performance checks. A site with a Largest Contentful Paint of 4 seconds is at a disadvantage compared to a competitor with 2 seconds. Mobile usability is another layer — Google uses mobile-first indexing, so if a page fails the mobile-friendly test, its ranking potential drops. For e-commerce sites with faceted navigation, audits catch duplicate content from URL parameters. I've seen sites with 100,000 indexed URLs because every filter combination created a unique URL. A proper audit flags that and suggests [canonical tags](/canonical-tag/) or noindex strategies.

I used to treat technical audits as a one-off project. I'd produce a comprehensive report, the client would fix the issues, and I'd move on. But sites change. New pages are added, plugins are updated, and redirects accumulate. I learned this the hard way when a client's site slowly lost rankings because a new developer removed the canonical tags from product pages. A recurring audit would have caught that within a week. Now I schedule quarterly checks and use automated monitoring tools to alert me when crawl errors spike. Another mistake: focusing only on obvious errors like 404s. I assumed that if Google could crawl the site, indexing would follow. But I missed indexation problems like pages that were in the crawl index but not in the search index — often due to thin content or soft 404s. I changed my approach after a site with 20,000 pages only had 5,000 indexed. A deeper dive revealed that the sitemap listed all pages but many returned a 200 with virtually no text. Now I use the URL Inspection Tool to sample pages and verify the rendered content. I also underestimated the impact of JavaScript on rendering. Early in my career, I relied on static HTML checks. Then a client using a React framework saw that their blog content was completely missing from Google's cache. The issue was that the content was loaded client-side and Googlebot timed out before rendering. Now I always test a few URLs with Google's Rich Results Test and check the HTML that Googlebot receives. The lesson: an audit is only as good as the depth you dig.