SEO Audit Guide: How to Audit Any Website in 2026

An SEO audit is the fastest way to find out why a website isn't ranking as well as it should. Done right, it surfaces the small number of real problems — usually fewer than you'd expect — that are holding the whole site back. Done wrong, it produces a 60-page PDF that nobody ships any fixes from.

This guide shows you how to do it right. We'll walk through the exact order of operations: what to check, in what sequence, with what tools, and how to decide which findings are actually worth your time. By the end you'll have a repeatable process that works whether you're auditing a 10-page consultancy site or a 100,000-page e-commerce catalogue.

What an SEO audit actually is

An SEO audit is a systematic review of the technical, on-page and content factors that determine how well a website performs in organic search. The goal isn't to produce a report. The goal is to produce a prioritised list of changes that, when shipped, will measurably improve rankings, traffic or both.

The mistake most people make is treating an audit as a diagnostic exercise instead of a strategic one. They list every issue they find, score them equally, and hand over a spreadsheet with 400 rows. Nobody reads spreadsheets with 400 rows. What actually moves the needle is identifying the three or five fixes that matter, explaining why they matter, and telling someone to ship them this sprint.

A good audit answers three questions:

1. Is Google able to find, crawl and index the important pages? If the answer is no, nothing else matters.
2. Are those pages actually signalling relevance for the queries we care about? This is on-page and content.
3. Is the site meeting the quality bar that Google now expects? Core Web Vitals, mobile experience, content depth, author signals.

When to run an SEO audit

There are four moments when an audit pays for itself:

• Before a redesign or migration. Catching URL structure problems, redirect gaps, or lost schema before launch is a hundred times cheaper than fixing them after traffic collapses.
• When rankings drop. A sudden dip usually traces back to one of a small set of causes — and an audit surfaces them faster than reading Twitter.
• Quarterly, as a baseline. Sites drift. New content gets published with bad titles, developers push changes that break canonical tags, images stop being compressed. A quarterly audit catches drift before it becomes debt.
• When you inherit a site. New job, new client, new acquisition — an audit is how you figure out what you actually have.

Before you start

Three things to have open before you run a single check:

• Google Search Console — verified for the site, with at least 28 days of data. This is your ground truth for what Google actually sees.
• A crawler — Screaming Frog, Sitebulb, or the crawl feature of Smart SEO Audit. Crawling is how you discover what's actually on the site, which is rarely what anyone thinks is on the site.
• Analytics — GA4, Plausible, or whatever the site uses. You need to know which pages actually get traffic, because those are the pages that deserve your audit time.

Decide on scope before you start. A full audit of a 50,000-URL site done manually is a week of work. If you only have two days, audit the top 500 pages by organic traffic — that's where 95% of the value lives anyway.

Step 1: Crawl the site

Start with a complete crawl. You want to see every URL the site exposes, every response code, every redirect chain, and every orphan page that's linked from nowhere but still responds to requests.

What to look for in the crawl data:

• 4xx errors on internal links — broken links waste crawl budget and hurt user experience
• Redirect chains longer than two hops — each hop dilutes link equity and slows page load
• Orphan pages — pages not linked from anywhere internally but still indexable
• Duplicate content — multiple URLs serving substantially identical content
• URLs with session IDs, tracking parameters, or faceted-navigation bloat
• Mixed content — HTTP resources loaded on HTTPS pages

Step 2: Audit indexability

If the right pages aren't indexed, or the wrong pages are, nothing else in this guide matters. Indexability is the foundation.

robots.txt

Check that robots.txt isn't blocking anything it shouldn't. A surprising number of sites accidentally block their entire product catalogue during a staging-to-production push. Fetch yoursite.com/robots.txt manually and read it line by line.

A clean robots.txt for most sites is short — a few disallow rules for admin areas, a sitemap reference, and nothing else.

User-agent: *
Disallow: /admin/
Disallow: /cart/

Sitemap: https://yoursite.com/sitemap.xml

Meta robots and X-Robots-Tag

Every important page should return index, follow either via a <meta name="robots"> tag or an HTTP header. Every thin, duplicate or utility page (tag archives, author pages, paginated filters) should be noindex — but verify that "noindex" isn't accidentally applied to your money pages.

Canonical tags

Canonicals tell Google which URL is authoritative when duplicates exist. Three rules:

1. Every indexable page should have a self-referencing canonical
2. Canonicals should point to the HTTPS, non-trailing-slash, lowercase version of the URL
3. A page set to noindex shouldn't have a canonical pointing elsewhere — that's a contradictory signal

XML sitemap

Your sitemap should contain only canonical, indexable URLs that return a 200 status code. It should not contain redirected URLs, 404s, or noindex pages. Submit it in Google Search Console and check the "Indexed, not submitted in sitemap" and "Submitted URL not indexed" reports weekly.

Step 3: Core Web Vitals and page speed

Google uses three metrics to measure the experience of actually loading a page. As of 2024, Interaction to Next Paint (INP) replaced First Input Delay (FID) as the responsiveness metric — make sure any tool you're using has been updated.

Pull field data (real user data) from Google Search Console's Core Web Vitals report or PageSpeed Insights. Lab data from Lighthouse is a useful second opinion but it's synthetic — field data is what Google actually uses for ranking.

Other speed checks

• Time to First Byte (TTFB) — under 600ms is good, under 200ms is excellent. Slow TTFB usually indicates a slow backend, unoptimised database queries, or a CDN that isn't caching aggressively enough.
• HTTP/2 or HTTP/3 — enabled on the server. Most modern hosts handle this by default, but old setups sometimes don't.
• Compression — Brotli or gzip on all text assets (HTML, CSS, JS).
• Image formats — WebP or AVIF for modern browsers, with appropriate fallbacks.
• Lazy loading — loading="lazy" on images below the fold.

Step 4: On-page SEO

For every important page — your top 20 to 100 URLs by traffic potential — check the on-page basics.

Title tag

The title should include the primary keyword, be under 60 characters to avoid truncation in SERPs, and read as something an actual human would click on. Front-load the keyword when it's natural, but don't torture the grammar for it. Google increasingly rewrites titles that feel stuffed.

Meta description

Not a ranking factor directly, but a significant click-through factor. Aim for 140-160 characters, include the target keyword once, and make the description describe what the user actually gets if they click. Generic descriptions cost you clicks.

H1

One H1 per page. It should contain or closely match the target keyword and describe the page's topic. Multiple H1s are technically valid in HTML5, but empirically sites with a single clear H1 perform better.

URL structure

Short, readable, keyword-containing URLs outperform long URLs with IDs and parameters. /guides/seo-audit beats /post?id=8742&cat=3. If the site already uses bad URL structures, fix them carefully — each URL change needs a 301 redirect to preserve link equity.

Heading hierarchy

H2s and H3s should outline the content logically. Avoid skipping levels (H1 → H3 with no H2). Headings help Google understand document structure and help users scan.

Image optimisation

Every content image needs descriptive alt text — not keyword-stuffed, just accurate. Alt text serves accessibility first; SEO benefit is a side effect of accessibility done right. Use descriptive filenames (seo-audit-checklist.png, not img_8723.png).

Internal links

Check that every important page is linked from at least three or four other pages using descriptive anchor text. "Click here" anchors are a wasted opportunity. "How to do an SEO audit" tells Google what the destination page is about.

Step 5: Content quality

Content quality is where most SEO audits are weakest, because it's the hardest thing to measure with tools. The checks here are manual, but they matter more than most of the technical stuff.

Thin content

Identify pages with under 300 words that aren't legitimately short (like product pages). These often need to be expanded, consolidated with related pages, or removed entirely.

Duplicate content

Use the crawler to find pages with similar titles, H1s, or body copy. Product variations, location pages, and tag archives are the usual culprits. Either canonicalise, consolidate, or differentiate.

Keyword cannibalisation

If two pages target the same query, neither ranks as well as one page would. Identify cannibalisation by pulling the "Pages" dimension in Search Console, filtering by query, and looking for queries where multiple URLs receive impressions. The fix is usually to merge the weaker page into the stronger one with a 301.

Outdated content

Pages ranking for time-sensitive queries lose traffic as they age. Content like "best SEO tools 2022" needs to become "best SEO tools 2026." Identify candidates by sorting your content by last-modified date and filtering for year-specific or news-type topics.

E-E-A-T signals

Experience, Expertise, Authoritativeness, Trustworthiness. Google doesn't score these directly, but they shape how it evaluates content. For Your-Money-Your-Life topics (health, finance, legal), check that articles have:

• A named author with a bio and credentials
• A clear publication and last-updated date
• Citations to primary sources
• Transparent about-us and contact pages

Step 6: Internal linking

Internal links do three things: distribute PageRank, signal content hierarchy, and help users navigate. A good internal linking audit finds pages that deserve more links, pages wasting link equity on unimportant targets, and structural weaknesses in the site architecture.

Metrics to review:

• Pages with fewer than 3 internal links — these are underpowered. If they matter, link to them more.
• Pages with no internal links (orphans) — Google finds these slowly if at all.
• Pages linked dozens of times with no organic traffic — link equity going to unproductive pages.
• Anchor text distribution — heavy use of "click here" or the page title as anchor, instead of descriptive phrases.

The highest-leverage internal linking change most sites can make is adding contextual links from high-traffic informational content (blog posts, guides) to the commercial pages that actually convert.

Step 7: Backlink profile

Backlinks are off-page, but an audit isn't complete without them. Use Ahrefs, Semrush, or Google Search Console's "Links" report to pull your referring domains.

What to check

• Lost links — domains that used to link to you and no longer do. Sometimes these can be recovered by contacting the site.
• Toxic links — spammy, paid, or low-quality links. Modern Google is mostly good at ignoring these, but if you have a lot of them, the disavow file still has a place.
• Link gaps — domains linking to competitors but not to you. These are the best outreach targets.
• Anchor text of inbound links — natural profiles have mostly branded and generic anchors. Heavy exact-match anchors look like link buying.

Internal signals from Search Console

GSC's "Top linking sites" and "Top linked pages" reports are free and underused. They tell you which external sites send links, and which of your pages attract links. Pages that attract lots of links but get little organic traffic often have on-page problems worth fixing — high link equity being wasted.

Step 8: Prioritise your fixes

By now you have a list of findings. A large list. The difference between an audit that gets shipped and an audit that gets filed is prioritisation.

For every finding, score two things on a scale of 1 to 5:

1. Impact — how much traffic or ranking improvement a fix likely produces
2. Effort — how much engineering or content work a fix requires

Plot them on a simple 2×2 matrix. High impact + low effort wins this week. High impact + high effort gets scheduled next quarter. Low impact + low effort is fine as a batch cleanup. Low impact + high effort gets cut from the list entirely.

Common SEO audit mistakes

Four things that consistently ruin otherwise-good audits:

1. Auditing everything equally

If you give the same attention to a page with 50,000 monthly visitors and a page with zero, the audit takes ten times as long and produces the same actionable output. Segment by traffic first.

2. Confusing symptoms with causes

"Rankings dropped" is a symptom. "We deployed a new template two weeks ago that broke the canonical tag" is a cause. Good audits trace symptoms back to causes instead of recommending fixes at the symptom level.

3. Prioritising by difficulty, not impact

It's tempting to start with the easy things. Sometimes the easy things are low-impact. A single well-rewritten title tag on a page with 10,000 monthly impressions often outperforms fifty technical tweaks.

4. Treating the audit as the deliverable

The audit isn't the deliverable. Shipped fixes are the deliverable. If the report sits in a drive folder without action items being assigned, the audit didn't happen.

Tools worth using

There are more SEO tools than anyone needs. This is what actually earns its keep:

You don't need all of them. GSC plus a good crawler plus an analytics tool is enough for most sites. The rest add speed, depth and automation.

Summary: the 20-minute audit

If you only have twenty minutes, do this:

1. Open Google Search Console. Check for manual actions and index coverage issues.
2. Run PageSpeed Insights on your three highest-traffic pages. Note failing Core Web Vitals.
3. Open your top 10 pages and verify each has a good title, meta description, H1, and canonical.
4. Search Google for site:yoursite.com and scan for unexpected URLs (staging pages, parameter URLs, etc.).
5. Check yoursite.com/robots.txt and yoursite.com/sitemap.xml for obvious problems.

That twenty minutes catches 60% of what a full audit catches. The remaining 40% is where a proper process — and a tool that runs every check for you — pays off.