What Should an SEO Audit Include? An Evidence Map
Scope the site, goals, dates, markets, search data, crawl and index paths, page fit, links, structured data, access, impact, owners, fixes, and recheck.
A useful SEO audit is not a long fault list. It links sound evidence to a user or search problem. It also names a fix, an owner, the likely impact, the cost, the risk, and a date to check again.
Set the Audit Frame
Write down the site scope, goals, markets, dates, changes, and key page types.
Save a baseline for search views, clicks, pages, actions, and sales fit.
Note data gaps, tool limits, access, and any parts you did not check.
Check Search and Page Paths
Test the crawl, index, status codes, robots, sitemap, canonical, and redirect paths.
Review mobile use, speed, security, access, markup, and page templates.
Match each query task to useful text, media, links, facts, and next steps.
Turn Findings Into Work
Show the evidence, affected pages, impact range, effort, risk, and owner.
Fix unsafe, broken, blocked, or high-value paths first. Small polish can wait.
Set a test, a release note, a check date, a stop rule, and a safe rollback.
Use the Technical SEO Checklist. Then prevent shelfware with Why Most SEO Audits Never Get Implemented.
Start With Scope and Business Tasks
Set the site, hosts, markets, and languages before the crawl. Add the dates, page types, users, and business tasks too. Define what the audit can check and what it cannot. Then save a baseline for search reach, good actions, sales fit, and known faults.
List every data source, owner, date range, limit, and access gap.
Sample page templates as well as high-value URLs.
Keep a change freeze or a change log while you measure the audit.
Cover the Technical Search Path
Check how a page is found, fetched, and rendered. Then check how it is indexed, chosen as canonical, linked, and shown. Review robots, sitemaps, status codes, redirects, canonicals, and hreflang. Also review JavaScript, mobile use, speed, security, structured data, and server logs when you have them.
Separate one-page faults from template and site-wide faults.
Test the final HTML and the user path, not just a tool score.
Show the affected URLs and one clean example for each finding.
Audit Intent, Content, and Proof
Map queries and internal search to the user's task and one best page. Check titles, headings, the direct answer, depth, facts, and dates. Also check authorship, sources, media, access, links, next step, and upkeep. Use people-first quality questions rather than a made-up E-E-A-T score.
Find overlap, thin pages, stale claims, missing proof, and content gaps.
Compare result types and competitors by task, not just word count.
Review the conversion and support paths after the search click.
Check Internal and External Signals
Review crawl paths, orphan pages, anchor text, related content, and broken links. Look at how useful internal links flow. Then review earned links and mentions for relevance, source, target, loss, and risk.
Do not call every weak or old link toxic.
Separate the links your team controls from the links it does not control.
Do not buy or place links to change rank.
Check entity names, profiles, citations, and local facts where they matter.
Use Tools as Evidence Collectors
A useful set may include Search Console, analytics, server logs, and a crawler. It may also include browser checks, speed tools, schema tests, rank samples, link data, and the CMS. Pick a tool for the question you have. Then record the limits of that tool.
Confirm a tool finding in the page, source, log, or official account.
Do not merge unlike dates, countries, devices, or metric rules.
Keep private exports secure. Remove them when the audit no longer needs them.
Deliver a Fix and Recheck Ledger
Each finding should show the issue, the evidence, and the affected set. It should also show the user and business effect, risk, effort, and dependency. Then name the owner, fix, test, release order, rollback, and recheck date. Group duplicates into one root cause.
Fix unsafe, blocked, broken, or high-value paths before small polish.
Give leaders a short choice list. Give implementers the exact evidence.
Re-crawl and compare the same sample after each release.
Close a finding only when the user path and search evidence pass.
Work a Priority Example
An audit finds 3,000 duplicate filter URLs and 40 key pages with wrong canonicals. It also finds 200 weak titles and one broken quote form. Fix the form and the wrong canonicals before you polish titles. Control the filter pattern at its source. Then recheck a small sample and the full affected set. These counts are a method example.
Priority one: user harm, blocked action, unsafe data, or lost access.
Priority two: crawl, index, canonical, template, and high-value page faults.
Priority three: content fit, proof, links, speed, and structured data.
Priority four: low-impact polish and ideas that still need evidence.
Do not close the root cause when only the sample URL was fixed.
The Short Answer
An SEO audit should show scope, evidence, limits, priority, owners, fixes, and rechecks. It should cover search data, crawl and index paths, and page fit. It should also cover links, markup, access, and impact. No audit can promise rank or traffic.
Need an SEO audit action map?
TTGC can map scope, evidence, issues, impact, effort, owners, tests, rechecks, and rollback. Search, product, data, legal, access, development, sales, and finance approval remain separate.
Sources
- Google Search Central: Google Search Essentials. https://developers.google.com/search/docs/essentials
- Google Search Central: SEO Starter Guide. https://developers.google.com/search/docs/fundamentals/seo-starter-guide
- Google Search Console Help: Inspect and troubleshoot a single page. https://support.google.com/webmasters/answer/12482179
- Google Search Console Help: Performance report. https://support.google.com/webmasters/answer/7576553
