“We need an SEO audit” usually means something different to whoever’s asking than whoever’s performing it. Here’s what a technical SEO audit actually checks — the specific, concrete items, not the vague version — so you know what you’re paying for and what a report should actually tell you.
Crawlability
Before anything else: can search engines actually reach your pages? This means checking robots.txt for accidental blocks, confirming important pages aren’t being disallowed by mistake, and reviewing crawl behavior for signs a search engine is wasting time on low-value pages (faceted navigation, duplicate parameter URLs) instead of the pages that matter.
Indexation
Being crawlable and being indexed aren’t the same thing. This step compares what’s actually indexed against what should be — flagging important pages that are missing from the index, and unimportant pages (staging duplicates, thin tag archives, internal search results) that shouldn’t be there but are, diluting the site’s overall quality signal.
Redirects
A redirect audit checks for redirect chains (A to B to C, when it should just be A to C), redirect loops, and old redirects that have quietly stopped working. This matters most after any past migration or URL structure change — it’s one of the most common sources of slow, invisible SEO decay.
Canonical tags
Canonicals tell search engines which version of a page is authoritative when duplicates exist — with or without trailing slash, with tracking parameters or without. Misconfigured or missing canonicals are a common, quiet cause of duplicate content issues that split ranking signal across multiple URLs instead of consolidating it.
XML sitemap
A sitemap audit checks that it exists, is submitted to search engines, and — just as important — that it accurately reflects the site: no dead pages, no missing important ones, and no non-canonical or noindexed URLs included by mistake.
Robots directives
Beyond robots.txt, this covers page-level noindex/nofollow tags — checking that they’re applied intentionally, not left over from a staging environment or applied broadly by a plugin default that nobody reviewed.
Site structure and internal linking
How pages connect to each other has a real effect on both crawl efficiency and ranking signal. This step looks at how many clicks it takes to reach important pages from the homepage, whether high-value pages are getting internal links from relevant content, and whether the site’s navigation and content genuinely reflect its topical priorities.
Page performance
Core Web Vitals and general load speed, checked against real thresholds rather than a generic “make it faster.” This usually surfaces specific, fixable culprits — unoptimized images, render-blocking scripts, a theme or page builder doing more work than the page needs.
Metadata
Title tags and meta descriptions, reviewed for missing pages, duplicates across multiple URLs, and titles that are either uninformative or stuffed with keywords in a way that reads badly to an actual person. This is one of the more visible fixes, and often one of the faster wins in a report.
Duplicate content
Beyond canonical issues, this checks for genuinely duplicated content across the site — product descriptions copied across variants, boilerplate text repeated on every location or service page — that dilutes uniqueness signals search engines use to evaluate content quality.
Broken links
Both internal (linking to pages that no longer exist) and external (linking out to pages that have since moved or disappeared) — a straightforward check that’s easy to overlook and cumulatively damaging to both user experience and crawl efficiency.
Migration-specific issues
If the site has been migrated, replatformed, or redesigned in the past, this is where those changes get checked specifically: were all old URLs redirected, did canonical tags survive the move, did the new structure preserve or fragment the internal linking that used to support important pages. This is a large enough topic on its own — we’ve covered recovering rankings after a bad migration in a separate piece.
Structured data
Where relevant, an audit checks whether structured data (schema markup for things like articles, products, or FAQs) is present, correctly formatted, and actually matches the visible page content. Structured data doesn’t guarantee better rankings on its own, but incorrect or mismatched markup can cause search engines to distrust it, and missing it entirely is a missed opportunity for the kind of enhanced search result listings that improve click-through rate.
Mobile usability
With mobile-first indexing the default for most sites, an audit checks the mobile rendering specifically — not just whether the site is “responsive” in a general sense, but whether tap targets, font sizes, and page structure hold up on an actual phone screen, and whether the mobile version of a page contains the same content and links as the desktop version. A site that looks fine on a laptop and is quietly broken on mobile is being evaluated by search engines primarily on the broken version.
How findings typically get prioritized
Not every issue an audit turns up deserves the same urgency. A useful way to sort findings is by two questions: how much traffic or ranking potential does this affect, and how much effort does it take to fix? A broken redirect on a high-traffic page is high impact and usually low effort — that’s a same-week fix. A full information-architecture overhaul might be high impact but high effort, and belongs in a longer-term plan rather than a launch checklist. Treating every finding as equally urgent is how audits turn into reports nobody acts on.
What happens after the audit
A technical audit is a diagnosis, not a treatment plan on its own. The useful next step is translating findings into a prioritized list of fixes, assigning realistic effort estimates, and — where the underlying cause is structural rather than a quick technical fix — connecting it back to decisions about site architecture, content strategy, or platform. An audit that stops at a list of problems is only half the job.
What a good audit report should give you
Not a wall of automated tool output — a prioritized list: what’s actually hurting rankings or traffic right now, what’s a smaller opportunity, and what’s genuinely not worth fixing yet. The output should be specific enough to act on, whether that action happens with us or with someone else entirely.
How this fits our process
A technical audit like this is typically the first concrete output of the Discover stage of any SEO engagement — the thing a written recommendation gets built from, not a sales document. If you’re not sure what shape an audit of your own site would actually take, let’s talk about it before assuming anything about scope.