A site audit is a review of a website's crawlability, indexability, technical SEO, content, performance, usability and business results.

Use **** as the audit date and record a baseline before making changes. The process below covers five tool groups, including Google Search Console, a crawler, Lighthouse or PageSpeed Insights, analytics and the Rich Results Test.

Site Audit at a Glance

Audit area What to check Main tools
Crawlability Robots.txt, XML sitemap, crawl errors Crawler, Search Console
Indexability Noindex tags, canonical URLs, indexed pages Search Console, URL Inspection
Technical SEO Status codes, redirects, HTTPS, duplicate URLs Crawler, browser tools
On-page SEO Titles, headings, descriptions, content relevance Crawler, manual review
Internal linking Broken links, orphan pages, important pages Crawler, analytics
Performance LCP, INP, CLS, mobile loading Lighthouse, Search Console
Structured data Valid and relevant schema markup Rich Results Test
Content quality Accuracy, originality, usefulness and search intent Manual review
Business results Leads, sales, calls and conversions Analytics, CRM

1. Define the Purpose and Scope of the Audit

Start by deciding what the audit needs to explain or improve.

Record:

  • The website's main business goals
  • Priority conversions, such as purchases, enquiries, calls or registrations
  • Important page types, such as product pages, service pages, category pages and blog posts
  • Main countries, languages and devices
  • Recent traffic or ranking declines
  • Recent migrations, redesigns, CMS changes or URL changes
  • The date range used for comparison

A small brochure website may need a page-by-page review. A large ecommerce website will usually need a technical crawl alongside manual reviews of representative templates.

Create a list of priority URLs before you start. Include the homepage, highest-value landing pages, top organic traffic pages, important conversion pages and pages that recently lost traffic.

2. Set Up the Audit Tools

Use at least these tools:

  1. Google Search Console for search performance, indexing, URL Inspection, Core Web Vitals and security notifications.
  2. A website crawler to collect URL-level data, including status codes, titles, headings, canonicals, directives and internal links.
  3. Lighthouse for page-level performance, accessibility, best practices and SEO checks. Lighthouse runs in Chrome DevTools, PageSpeed Insights, from the command line or as a Node module.
  4. Google Analytics or another analytics platform for engagement and conversion data.
  5. Rich Results Test for validating eligible structured data.

Export the current data before making changes. This creates a baseline for measuring changes in crawling, rankings, traffic, speed and conversions.

3. Crawl the Website and Build a URL Inventory

Run a crawl that includes the website's main HTML pages, images, PDFs and other important assets.

For every URL, collect:

  • HTTP status code
  • Indexability status
  • Canonical URL
  • Title tag
  • Meta description
  • H1 heading
  • Word count or visible content
  • Internal links in and out
  • Images and missing alt text
  • Redirect destination
  • Robots directives
  • Structured data
  • Page depth from the homepage

Compare the crawl with:

  • URLs in the XML sitemap
  • URLs shown in Google Search Console
  • URLs receiving organic traffic
  • URLs receiving conversions
  • URLs linked from important pages

This comparison can reveal pages that are published but not linked, pages receiving traffic but missing from the sitemap, and sitemap URLs that should not be indexed.

Check for These Crawl Problems

  • Broken internal links
  • 4xx and 5xx errors
  • Redirect chains
  • Redirects pointing to irrelevant pages
  • Multiple URL versions for the same content
  • Orphan pages with no internal links
  • Pages that are too deep in the site structure
  • Accidental staging or test URLs
  • Filter, sort and search URLs creating duplicates
  • HTTP pages that should resolve to HTTPS

A crawler warning is not automatically a priority. It matters when it affects a valuable page, many URLs or an important user journey.

4. Check Whether Google Can Crawl and Index Important Pages

Crawling and indexing are separate processes. Crawling means Google can request a URL. Indexing means Google has processed the page and stored it as a candidate for search results.

Review Robots.txt

A robots.txt file controls whether crawlers can access particular URLs. It does not reliably remove a page from Google's index. A URL blocked by robots.txt can still appear in search if Google discovers it elsewhere.

Use noindex, authentication or page removal when the goal is to prevent indexing.

Check that robots.txt does not block:

  • Important page templates
  • CSS or JavaScript needed to render the page
  • Images required for image search
  • Product, category or service pages
  • URLs needed to understand internal navigation

Review the XML Sitemap

An XML sitemap should contain the preferred, canonical URLs that you want Google to discover and index. A sitemap is a signal, not a guarantee that Google will crawl or index every listed URL.

Google recommends using absolute URLs and keeping the sitemap focused on canonical pages.

Remove URLs that are:

  • Redirected
  • Non-canonical
  • Blocked from indexing
  • Returning errors
  • Thin, expired or intentionally excluded

Use Search Console URL Inspection

Inspect your most important URLs individually. Check:

  • Whether Google can access the page
  • Whether the page is indexed
  • Which canonical URL Google selected
  • Whether the page has indexing or rendering problems
  • The last available crawl information

The Page Indexing report shows indexed and non-indexed URLs. A non-indexed URL is not automatically a problem. Duplicate, alternate or intentionally excluded pages may be correctly absent from the index.

The question to answer is whether the correct canonical version of each valuable page is indexed.

5. Audit Redirects, Canonical URLs and Duplicate Pages

Every important page should have one clear preferred URL.

Check for:

  • HTTP and HTTPS versions
  • www and non-www versions
  • Trailing slash variations
  • Uppercase and lowercase URL variations
  • URL parameters
  • Printer or mobile versions
  • Duplicate category and product URLs
  • Similar regional or language pages
  • Canonical tags pointing to the wrong page
  • Redirects pointing to unrelated content

Google treats redirects as a strong canonicalisation signal, rel="canonical" as a strong signal and sitemap inclusion as a weaker signal. These signals should agree wherever possible.

Robots.txt should not be used as a canonicalisation method.

For each duplicate group, choose one of these actions:

  • Keep one URL and permanently redirect the alternatives
  • Add a correct canonical tag
  • Use noindex when the page is useful to users but not useful in search
  • Improve the pages so they serve clearly different purposes
  • Remove obsolete pages

Do not automatically canonicalise pages that target different search intents. Two pages may look similar technically but still deserve separate URLs if they serve different audiences, products or locations.

6. Review Technical On-Page SEO

Inspect the page elements that help users and search engines understand each URL.

Title Tags

Every indexable page should have a unique, descriptive title that accurately reflects the page content. Google generally uses the HTML <title> element as one source for the title link shown in search results, although it may also use other page text or links.

Look for:

  • Missing titles
  • Duplicate titles
  • Titles that are too vague
  • Titles overloaded with keywords
  • Titles that describe a different page
  • Important terms placed too late
  • Brand names repeated unnecessarily

Meta Descriptions

A meta description should summarise the specific page and give searchers a reason to click. Google may use the meta description, but it can generate a different snippet when another section better matches the query.

Check for:

  • Missing descriptions on important pages
  • Duplicate descriptions
  • Descriptions that do not match the page
  • Descriptions with no clear value proposition
  • Descriptions written as keyword lists

Headings and Page Structure

Check that each page has:

  • One clear primary heading
  • Logical subheadings
  • A structure that matches the searcher's task
  • Important information visible in the main content
  • No headings used only for visual styling

A service page should explain the service, audience, process, benefits, limitations and next step. A product page should make the product, specifications, price, availability and purchase action easy to find.

Images

Review:

  • Missing or misleading alt text
  • Oversized images
  • Images that cause layout shifts
  • Important text embedded only in images
  • Broken image URLs
  • Decorative images incorrectly treated as meaningful content

Alt text should describe the image's purpose in context. It should not be a list of keywords.

7. Audit Content Quality and Search Intent

A technically sound website can still perform poorly when its content does not satisfy the query.

For each important page, ask:

  1. What exact question or task does this page address?
  2. Does the page answer that question immediately?
  3. Is the content more useful than competing results?
  4. Does the page provide original information, examples, evidence or analysis?
  5. Is the information accurate and current?
  6. Does the page demonstrate relevant expertise?
  7. Is the next step clear?
  8. Does the page overlap with another URL on the site?

Google recommends people-first content that is useful to the intended audience, demonstrates expertise and provides substantial value rather than content created mainly to attract search traffic.

Classify each page as:

  • Keep: useful, relevant and performing well
  • Improve: valuable topic but weak content, structure or conversion path
  • Consolidate: overlaps substantially with another page
  • Redirect: obsolete page with a suitable replacement
  • Remove or noindex: thin, redundant or intentionally unavailable content

Do not delete pages solely because they have low traffic. A page may support internal linking, assist conversions, target a low-volume commercial query or serve returning customers.

Internal links help users navigate and help search engines discover relationships between pages.

Check that:

  • Every important page has internal links
  • Important pages are reachable within a reasonable number of clicks
  • Navigation reflects the site's main categories
  • Anchor text describes the destination
  • Links do not point to redirected or broken URLs
  • Related content is connected naturally
  • High-value pages receive links from relevant pages
  • No important page depends only on the XML sitemap for discovery

Pay particular attention to orphan pages. An orphan page may be indexed or receive external links, but it is difficult for users and crawlers to find through the site itself.

9. Test Performance and Core Web Vitals

Run Lighthouse on representative templates rather than only the homepage. Test a homepage, service page, product page, category page, article and checkout or lead form where applicable.

Google's Core Web Vitals measure loading performance, responsiveness and visual stability:

Metric What it measures Good target
LCP Loading performance 2.5 seconds or less
INP Responsiveness to interactions Less than 200 milliseconds
CLS Visual stability Less than 0.1

These targets are assessed at the 75th percentile and should be considered separately for mobile and desktop users.

Common causes of poor performance include:

  • Large, uncompressed images
  • Excessive JavaScript
  • Slow server response times
  • Render-blocking resources
  • Third-party advertising or tracking scripts
  • Web fonts that delay visible text
  • Missing image dimensions
  • Layout changes caused by banners or pop-ups

Lighthouse provides lab diagnostics, but lab results are different from real-user data. Interaction to Next Paint requires user interaction and cannot be fully measured in a simple page-load lab test.

Compare Lighthouse findings with the Core Web Vitals report in Search Console where data is available.

10. Check Structured Data

Structured data helps Google understand entities and page types. It can make a page eligible for rich results, but valid markup does not guarantee that enhanced search features will appear.

Audit whether:

  • The schema type matches the page
  • The marked-up information is visible to users
  • Required properties are present
  • Product prices and availability are current
  • Reviews represent genuine visible content
  • Multiple plugins are generating conflicting markup
  • Structured data passes the Rich Results Test
  • Search Console reports new enhancement errors after deployment

Prioritise structured data for pages where richer search results could improve qualified clicks, such as products, recipes, events, local businesses and software applications.

A site audit should also examine factors that affect trust and conversions.

Check:

  • HTTPS across all important URLs
  • Security warnings or hacked-content notifications
  • Forms and checkout processes
  • Mobile navigation
  • Cookie and consent behaviour
  • Accessibility barriers
  • Contact details and business information
  • Privacy, returns and delivery information
  • Intrusive interstitials
  • Broken telephone, email and booking links
  • Analytics and conversion tracking

Lighthouse includes accessibility and best-practice audits alongside performance and SEO checks. Use those results as a starting point, then manually test important user journeys because automated checks cannot identify every usability problem.

12. Prioritise the Fixes

An audit should lead to a short action plan rather than a long list of warnings.

Use this prioritisation model:

Priority Typical issue Action
Critical Important pages blocked, site offline, hacked content, incorrect migration redirects Fix immediately
High Valuable pages not indexed, major canonical errors, broken templates, severe mobile problems Fix next
Medium Duplicate titles, weak internal links, slow images, missing schema Schedule by template
Low Minor metadata inconsistencies or isolated warnings on low-value pages Fix during routine maintenance

For each issue, record:

  • URL or URL pattern
  • Problem
  • Evidence
  • Business or SEO impact
  • Recommended fix
  • Person responsible
  • Effort required
  • Priority
  • Validation method

When choosing between competing fixes, compare:

  • Expected impact
  • Number of affected URLs
  • Implementation effort
  • Confidence that the fix will solve the problem

13. Re-Crawl and Measure the Results

After implementation:

  1. Re-crawl the affected URLs.
  2. Confirm that redirects, canonicals and indexability directives are correct.
  3. Use URL Inspection for high-value pages.
  4. Validate structured data again.
  5. Re-run Lighthouse on the same templates.
  6. Monitor Search Console impressions, clicks, indexing and Core Web Vitals.
  7. Check analytics and conversion data.
  8. Compare performance with the original baseline.

Search Console performance data includes clicks, impressions, click-through rate and average position. Trends in clicks and impressions are generally more useful for measuring search growth than average position alone.

Final Site Audit Checklist

Before closing the audit, confirm that:

  • Important pages are crawlable
  • Important pages are indexable
  • Canonical URLs are correct
  • XML sitemaps contain preferred URLs
  • Robots.txt does not block valuable resources
  • Broken links and redirect chains are fixed
  • Titles, headings and descriptions match search intent
  • Important pages have internal links
  • Duplicate or obsolete pages have a clear treatment
  • Content is accurate, useful and differentiated
  • Core Web Vitals problems are assigned to an owner
  • Structured data is valid and representative
  • HTTPS and key user journeys work correctly
  • Fixes have been re-crawled and measured

The most effective site audit produces a short, prioritised implementation plan, not a large automated score. Start with pages that drive revenue or leads, fix crawl and indexation barriers first, then improve content, internal linking, performance and conversion paths.