Technical SEO is not a checklist of tags. It is making sure a search engine can reach your pages, render them, understand what they are for, and index the right one. Content strategy cannot compensate for a site that blocks any of those.
What we examine
- status codes, redirects and redirect chains
- robots.txt, meta robots and X-Robots-Tag headers
- XML sitemaps and what they include or omit
- canonical tags and self-reference consistency
- JavaScript rendering and what appears without it
- internal linking and orphaned pages
- structured data validity
- Core Web Vitals in field data, not just lab scores
- hreflang on multilingual sites
- crawl budget, facets and parameters on large stores
Findings are ranked, not listed
An audit that hands over four hundred issues has moved the problem rather than solved it. We sort by impact, risk, effort and dependency, so the first week of work is the week that matters.
Developers get tickets naming the affected template, an example URL, the expected behaviour and acceptance criteria. Then we re-check after deployment, because "fixed" and "deployed and actually working" are different states.
The failures that cost the most
In practice, most serious technical damage comes from a handful of causes: a relaunch without redirect mapping, a noindex left on after a staging deploy, canonical tags pointing every page at the home page, or a JavaScript build that renders nothing without execution.
Each is invisible from the front end and expensive to leave running. That is why measurement comes before opinion.
Duplicate and near-duplicate URLs
The same page reachable at several addresses splits its own signals. Common causes: http and https both serving, www and non-www both answering, trailing-slash variants, session parameters, and filtered category views generating thousands of thin combinations.
The fix is one canonical form, permanent redirects to it, and a clear rule for what may be indexed.
Core Web Vitals
Field data from real visitors matters more than a lab score on a fast connection. We look at what actually delays the largest paint, what shifts layout after load, and what blocks interaction — then trace each to a decision in the build.
FAQ
How long does an audit take?
Usually one to three weeks depending on the size of the site and access to data.
Do you implement the fixes?
Yes, where that is agreed. Otherwise your developers get tickets specific enough to act on.
Will this improve rankings on its own?
Technical work removes obstacles. If the content does not answer the query, removing obstacles is not enough.