Perform an SEO website audit by defining scope, collecting crawl and Search Console evidence, reviewing representative pages and turning confirmed findings into a prioritised action plan. Keep observations, hypotheses and expected outcomes separate so each fix can be retested.

Audit time depends on scope, access, crawl size and the number of templates or markets sampled. Record those limits before collecting data so the final report does not imply full coverage when the review used a sample.

Step 1: Set Up Your Audit Tools

Before you touch a single page, you need the right toolkit in place. Trying to audit a website without proper tools is like trying to diagnose a car engine by listening to it from across the street. Here is what you need at minimum:

Google Search Console — free, essential. Gives you real indexation data, search queries, click-through rates, and Core Web Vitals from Google's perspective.
A site crawler — Screaming Frog (free mode currently up to 500 URLs), Sitebulb, or Lumar. Crawlers inventory URLs reachable from their seeds and configuration; reconcile results with sitemaps, Search Console and known platform URLs.
Google Analytics or equivalent — for traffic trends, bounce rates, and conversion data that add context to ranking data.
A backlink tool — Ahrefs, Semrush, or Moz. You need one to analyse your link profile and compare against competitors.
PageSpeed Insights — free from Google. Tests Core Web Vitals with both lab and field data.

Create a dedicated folder or spreadsheet to collect your findings as you move through each step. Label columns for the issue, affected URL, severity (critical, high, medium, low), and recommended fix. This becomes your action plan at the end.

Step 2: Crawl Your Website

Your first active step is running a complete crawl of your website. Configure your crawler to follow internal links, respect robots.txt (or ignore it if you want to see what is blocked), and store response codes for every URL.

Pay attention to these crawler outputs:

HTTP status codes — confirm successful pages, redirects, client errors and server errors. Trace chains and loops to their final response; prioritise them by affected links and user impact rather than a fixed hop threshold.
Redirect chains and loops — a common problem after site migrations. Every chain should resolve in a single hop.
Orphan pages — URLs in your sitemap or Google index that have zero internal links pointing to them. These pages are almost invisible to search engines.
Duplicate content — pages with identical or near-identical content, often caused by URL parameters, trailing slashes, or HTTP/HTTPS variants.
Crawl depth — important pages should be reachable within three clicks from the homepage. Anything deeper gets crawled less frequently.

Export the full crawl data. You will reference it repeatedly in later steps. Most crawlers let you export to CSV or Excel, which makes filtering straightforward.

Step 3: Check Indexation Status

Open Google Search Console and navigate to the Pages report (formerly Index Coverage). This tells you exactly how many of your pages Google has indexed, and more importantly, why certain pages have been excluded.

Compare the number of indexed pages in GSC against the total URLs found by your crawler. A large gap means Google is choosing not to index a significant portion of your site. Common reasons include:

Crawled, currently not indexed — Google reports a crawl but no current index inclusion. The label does not by itself prove a content-quality decision.
Discovered, currently not indexed — Google knows about the URL but has not indexed it. Inspect representative URLs and related crawl evidence before assigning a cause.
Blocked by robots.txt — you are accidentally preventing Google from accessing important pages.
Noindex tag — a meta robots noindex directive is present. Sometimes added intentionally, sometimes left over from a staging environment.
Duplicate, Google chose a different canonical — Google thinks another URL is the primary version and has ignored your stated canonical.

For every category, investigate the affected URLs. Cross-reference with your crawl data. If Google is not indexing pages you consider important, you need to understand why and fix the root cause before worrying about rankings.

Step 4: Analyse On-Page SEO

Review the intended landing pages for important search tasks and sample shared templates. On-page changes can clarify relevance and improve snippets, but their effect depends on the query and competing results.

Title tags — check that each title is accurate, descriptive and distinct for its page. Review actual title links because Google can generate them from several on-page sources and displayed length varies.
Meta descriptions — write an accurate page summary for likely visitors. Google may use the description or other page text for the snippet, so a fixed length or call to action is not a ranking requirement.
H1 and headings — use a clear primary heading and a logical hierarchy. Multiple H1 elements are valid HTML; flag them only when they make structure or page purpose unclear.
Heading hierarchy — H2s, H3s, and H4s should follow a logical outline. Skipping levels (H1 straight to H3) signals poor content structure.
Query alignment — confirm that the page answers its intended search task in the opening and subordinate sections. Avoid fixed keyword placement or density rules.
Internal links — every important page should receive internal links from related content. Check anchor text distribution. Generic anchors like "click here" waste link equity.
Image alt text — descriptive, keyword-relevant alt attributes help with image search and accessibility. Missing alt text is one of the most common audit findings.

Create a spreadsheet row for each page with the current title, current H1, target keyword, and notes on what needs changing. This becomes the on-page section of your action plan.

Step 5: Review Technical SEO

Technical SEO ensures search engines can efficiently crawl, render, and index your content. Many technical issues are invisible to users but devastating to rankings. Check these areas:

Robots.txt: Open your robots.txt file (yourdomain.com/robots.txt) and verify it is not blocking important directories. A surprisingly common mistake is blocking CSS and JavaScript files, which prevents Google from rendering your pages correctly.

XML Sitemaps: Your sitemap should list every indexable page and exclude pages with noindex tags, redirects, or 404 errors. Check that the sitemap is referenced in robots.txt and submitted in Google Search Console. The last-modified dates should be accurate, not set to today's date on every page.

Canonical tags: Every page should have a self-referencing canonical tag unless it is a duplicate that points to a primary version. Inspect your canonical tags for common errors: relative URLs instead of absolute, HTTP instead of HTTPS, or pointing to a non-existent page.

Structured data: select markup that represents visible page content and validate Google-supported types with the Rich Results Test. This page retains HowTo vocabulary so other consumers can identify the visible ten-step procedure, but Google no longer shows HowTo rich results and limits FAQ rich results mainly to well-known government and health sites.

HTTPS: Every page should load over HTTPS. Mixed content warnings (loading HTTP resources on an HTTPS page) can break trust signals and display browser warnings to users.

Hreflang (if multilingual): Check that hreflang annotations are reciprocal, use correct language-region codes, and include a self-referencing tag. Hreflang errors are among the most difficult to debug.

Step 6: Evaluate Content Quality

Google's people-first content guidance asks whether content is useful, reliable and made primarily for people. Evaluate the answer, evidence and trust in context rather than assigning a generic quality score.

Identify content that is inaccurate, duplicative or unable to complete its intended task. Word count alone does not define thin content. Expand, combine or remove a page only after reviewing its purpose, links, demand and required redirects.

Next, investigate possible query overlap. Multiple URLs for one query can be legitimate when they serve different intents. Compare page purpose, query variants, trends, canonicals, links and conversions before deciding whether to differentiate, consolidate or leave them alone.

Check for content freshness. Pages with outdated statistics, broken outbound links, or references to past years (like "best tools for 2023") signal neglect. Update them or add a visible "last updated" date.

Finally, assess topical coverage. Map your content against the topics your audience cares about. If competitors cover subtopics that you have ignored entirely, you have a content gap that limits your topical authority.

Step 7: Audit Your Backlink Profile

External links can help discovery and are one source of authority evidence. Review referring pages, context, acquisition history and lost links without treating a domain count or third-party toxicity label as a ranking score.

Export your full backlink profile from Ahrefs, Semrush, or Moz. Analyse the following:

Total referring domains — more important than total backlinks. One link from 100 different sites is stronger than 100 links from one site.
Domain authority distribution — a natural profile has links from sites of varying authority. An unnatural spike of high-DR links can trigger scrutiny.
Anchor context — review repeated commercial anchors and the pages that created them. Do not use a universal percentage to label a profile manipulative.
Link risk — investigate backlinks connected to deliberate spam or a manual action. Treat vendor toxicity scores as leads for review rather than Google's assessment.
Lost links — valuable backlinks that have disappeared. If a high-authority site removed your link, investigate whether you can reclaim it through outreach.
Competitor comparison — compare your link profile against the top three competitors for your primary keywords. Identify domains that link to them but not to you. These become outreach targets.

Use Google's disavow tool only when a substantial pattern of artificial links and a manual action or likely manual action justify it. Make reasonable removal efforts first.

Step 8: Check Site Speed

Performance can affect page experience and task completion, while Core Web Vitals are one input among many search signals. Use the site's field data and analytics to test whether a measured delay is associated with abandonment; do not infer a bounce-rate change from a lab score alone.

Run your homepage and five to ten key landing pages through PageSpeed Insights. Record these metrics:

Largest Contentful Paint (LCP) — should be under 2.5 seconds. This measures when the main content finishes loading. Common causes of poor LCP: unoptimised hero images, slow server response times, render-blocking CSS or JavaScript.
Interaction to Next Paint (INP) — should be under 200 milliseconds. Measures responsiveness when users interact with the page. Heavy JavaScript frameworks and long-running tasks are the usual culprits.
Cumulative Layout Shift (CLS) — should be under 0.1. Measures visual stability. Ads, images without dimensions, and dynamically injected content cause layout shifts.
Total Blocking Time (TBT) — use this lab diagnostic to find main-thread blocking. Interpret the score against the current tool version, then profile long tasks before changing JavaScript.

Use WebPageTest for deeper diagnostics. It shows a waterfall chart of every resource request, making it easy to identify the specific files slowing your page. Common fixes include enabling compression, setting cache headers, lazy-loading below-the-fold images, and removing unused CSS and JavaScript.

Step 9: Assess Mobile Experience

Google uses mobile-first indexing, meaning it evaluates the mobile version of your site for ranking purposes. A site that looks perfect on desktop but breaks on mobile will underperform in search.

Test your site on real devices, not just Chrome DevTools. Check for:

Responsive layout — does the design adapt properly at 375px, 414px, and 768px widths? Content should not overflow horizontally.
Tap targets — test whether controls are distinguishable and operable without accidental activation. Use the accessibility standard selected for the audit rather than a claimed Google ranking threshold.
Text readability — test readable sizing, spacing, contrast and zoom on supported viewports. A fixed 16 px value cannot account for typeface, spacing and user settings.
Interstitials and overlays — test whether mobile users can immediately access the main content and complete the page task. Google's guidance allows some legally required notices and discourages intrusive promotions, but eligibility depends on implementation; do not label every consent banner exempt or every modal a penalty.
Viewport configuration — ensure your pages include a proper viewport meta tag. Without it, mobile browsers render the page at desktop width and shrink it.
Mobile content parity — the mobile version should contain the same content as desktop. Hidden tabs, accordions that collapse content, or "read more" truncation can cause Google to miss important text.

Document every mobile issue you find along with the affected page and a screenshot. Mobile fixes often require CSS changes that development teams can batch together.

Step 10: Create Your Action Plan

An audit without an action plan is just a list of problems. The final step is organising your findings into a prioritised roadmap that your team can execute.

Sort every finding into one of four priority levels:

Critical — confirmed issues that block intended indexing, break primary journeys or create an urgent security risk. Assign an owner and deadline based on business impact.
High — confirmed issues with material reach or user impact. Document dependencies and choose a deadline the delivery team can meet.
Medium — useful improvements with lower reach, confidence or urgency. Schedule them after higher-impact work.
Low — observations with limited impact or optional refinements. Monitor or batch them when efficient.

For each issue, document the affected URLs, the specific fix required, the evidence and the verification method. Assign an owner and review date based on severity, release timing and how long the relevant system needs to process the change.

Repeat the checks when releases, migrations, unexplained search changes or the site's risk profile justify them. Use the audit frequency guide to define owners and triggers, then remeasure the same evidence after implementation.

How to Do a DIY SEO Website Audit in 10 Steps