What is a site audit?
Your website is the place buyers, search engines and AI assistants all check before anyone fills out a form. A site audit tells you which problems on it are costing you leads, so you fix those first.
How a site audit works
An audit starts with a crawl. A tool such as a desktop crawler or an SEO platform visits your pages the way Googlebot does, follows every internal link and records what it finds: status codes, titles, canonical tags, load times, word counts and schema markup.
Some audits stop at the crawl. On its own, a crawl is the least useful part. A crawl report treats a broken link on a 2019 press release the same as a broken link on your top product page. The real work is the second step: matching each issue to the pages that bring in leads and revenue, then ranking the list.
Say a crawl flags 300 issues on a manufacturer’s site. Twelve of them sit on the five pages that drive most quote requests, including a redirect chain on the main product category and a missing canonical on a spec sheet. Those twelve go to the top. The other 288 get batched into a cleanup sprint or parked.
A full audit usually covers three layers:
- Technical: crawlability, indexation, redirects, page speed, mobile rendering, structured data and XML sitemaps.
- Content: thin or duplicate pages, keyword cannibalization, outdated pages and gaps against what buyers search for.
- Visibility: how pages rank, which pages earn clicks, and whether AI assistants can read and cite the site.
The six parts of a site audit
Crawl and index
Can search engines reach every page that matters? Look for noindex tags on money pages, orphan pages, robots.txt blocks and sitemap errors.
Redirects and errors
404s, redirect chains and links that point at redirects waste crawl budget and leak link equity. Fix each link at its source.
Speed and experience
Core Web Vitals, mobile layout and render-blocking scripts. Slow pages lose buyers before the form loads.
On-page content
Titles, meta descriptions, headings, duplicate copy and pages that compete for the same query.
Structured data
Schema markup tells search engines and AI systems what a page is about: a product, a service, an FAQ, an organization.
AI crawler access
Whether GPTBot, ClaudeBot, PerplexityBot and Google-Extended can read your pages, and whether your core answers live in plain HTML.
How to prioritize findings by revenue impact
Every finding gets two scores. The first is revenue impact: does the issue touch a page that ranks, converts or feeds the sales team? The second is effort: is it a five-minute settings change or a developer sprint?
To score impact, pull the pages that earn organic clicks from Google Search Console and the pages that produce form fills or orders from your analytics and CRM. Any issue on those pages moves up. Issues on pages with no traffic and no conversions move down, however alarming the crawler makes them sound.
Why audits now include AI crawler access
ChatGPT, Perplexity, Claude and Google’s AI features answer buyer questions by fetching and reading web pages. If your robots.txt blocks their crawlers, or your product details only load through JavaScript, those systems have nothing to quote. Your competitor’s page gets cited.
So a current audit adds a few checks: which AI user agents are allowed, whether core answers appear in the raw HTML, whether you publish an llms.txt file, and whether your entity details (name, address, products) match across the web. You can run a quick version of this with our free AI visibility checker, and read how AI search works for the background.
From our work
Here’s how we’d approach an audit for an industrial manufacturer that sells through quote requests. We’d start with the RFQ path: the product category pages, the spec sheets and the quote form. Every crawl issue on those pages gets fixed first, even small ones, because a broken redirect there costs a sale.
Next we’d check AI access: robots.txt rules for each AI crawler, whether spec tables render in plain HTML, and whether schema describes the products. Then we’d fix internal links at their source after any slug changes, so no money page sits behind a redirect. Everything else goes into a cleanup backlog with an owner and a date. For the full scope, see our technical SEO service.
A quick site audit checklist
- Crawl the whole site and export every issue with its URL.
- Pull clicks by page from Search Console and conversions by page from analytics.
- Tag each issue with impact (money page or not) and effort (minutes, hours, sprint).
- Check robots.txt for AI crawlers and confirm core content is in the HTML.
- Fix the high-impact, low-effort list this week.
- Re-crawl after the fixes to confirm they held, then repeat each quarter.
Related terms
A site audit touches most of technical SEO. Start with Technical SEO, crawlability and indexation, then read up on redirect chains, Core Web Vitals, robots.txt, AI crawlers and the site health score most audit tools report. Our SEO services and AI SEO services pages show where audits fit in a full program.
Frequently asked questions
What is included in a site audit?
A site audit covers technical health (crawling, indexing, redirects, speed, mobile, structured data), content quality (duplicate, thin and outdated pages) and visibility (rankings, clicks and AI crawler access). A good one ends with a ranked fix list tied to revenue pages.
How long does a site audit take?
The crawl itself takes minutes to hours depending on site size. The analysis, where you map issues to traffic and conversion data and rank the fixes, takes longer. Plan for days on a large ecommerce or multi-language site.
How often should you do a site audit?
Run a full audit once a year and a lighter crawl every quarter. Also audit after any site migration, redesign, platform change or big drop in organic traffic.
What is the difference between a site audit and an SEO audit?
The terms overlap. “SEO audit” often means the search-ranking parts only. A site audit usually takes a wider view that includes speed, user experience, analytics tracking and, today, AI crawler access.
Fix what costs you leads first
We audit sites for Google and AI search, then rank every fix by its likely effect on pipeline and revenue.
