SEO Audit Checklist: 9 Easy Steps to Find Costly Problems

Share On:

SEO audit process showing the 9 steps from crawling to prioritized fixes

An SEO audit is a structured check of whether search engines can find, index, understand, and trust your pages, followed by a ranked list of what to fix first. The ranked list is the part that matters. Any crawler can produce 300 warnings in five minutes. An audit tells you which three of them are costing you traffic.

To run one, work in this order: confirm Google can crawl your site, check what it actually indexed, crawl the site yourself, then review page experience, content, links, and AI search visibility. Each stage can block everything after it, so a problem early in the chain outranks any number of problems later on. Below are the nine steps we follow, the free tools that cover most of them, and a worked example of turning findings into a fix list.

What An SEO Audit Is, And What It Is Not

A real audit answers one question: what is stopping this site from getting the organic traffic its content deserves? It covers technical health, content, authority, and, since 2024, how the site shows up in AI answers. It ends with recommendations ordered by impact, not a list sorted by the tool’s severity color.

It is not a PDF of every warning Semrush or Ahrefs raised. Those exports are a starting point, and a useful one. But a tool doesn’t know that your “missing meta descriptions” sit on 150 tag archives nobody should index, or that the one noindex tag it flagged is on your main services page. Judgment is the part you’re paying for, or the part you have to supply yourself.

Google Search Console page indexing report used in an SEO audit

 

Why The Order Of An SEO Audit Matters More Than The Checklist

Search visibility works like a chain. If a page can’t be crawled, it can’t be indexed. If it isn’t indexed, its content quality is irrelevant. If the content doesn’t match what people search for, backlinks won’t save it. So we audit in stages and fix in the same order.

Stage The question Typical blocker
Find Can search engines reach the page? robots.txt disallow, broken internal links, orphan pages
Index Did Google choose to index it? noindex tag, wrong canonical, thin or duplicate content
Render and load Can it load and display properly for real users? Slow LCP, content injected only by JavaScript, mobile layout problems
Understand Does it clearly answer a real query? Intent mismatch, two pages competing for one query, vague titles
Trust Does the wider web vouch for it? Few relevant links, inconsistent business details, no visible expertise
Cite Do AI answers use it? AI crawlers blocked, key facts buried, no clear answers to questions

 

When you rank findings later, a Find or Index problem on an important page almost always beats anything further down the table.

How Big Should Your SEO Audit Be?

A 40 page plumbing site and a 60,000 URL ecommerce store need different audits. Scaling the scope is where most checklists go wrong.

Site size Focus on Usually skip or park
Under 500 URLs (most local and service businesses) Indexing status of every key page, titles and intent, Google Business Profile, Core Web Vitals on templates, internal links to money pages Crawl budget, log file analysis, faceted navigation
500 to 10,000 URLs (content sites, small ecommerce) Duplicate and parameter URLs, category and product templates, cannibalization, internal link depth, content pruning Crawl budget, unless GSC shows many “Discovered” URLs
10,000+ URLs, or content that changes daily Crawl budget, log files, faceted navigation, pagination, hreflang if international, template level fixes Page by page content review (sample by template instead)

 

The crawl budget line comes from Google itself. Its crawl budget guide is written for sites with 1 million+ unique pages that change weekly, sites with 10,000+ pages that change daily, or sites with a large share of URLs stuck in “Discovered – currently not indexed”. If your pages get crawled the day you publish them, Google says you don’t need to read it.

What Tools Do You Need For An SEO Audit?

Less than you think. A free stack covers most of the work on a small site:

  • Google Search Console for indexing, Core Web Vitals field data, search queries, links, and the Generative AI performance report.
  • Google Analytics 4 to see which pages actually bring organic sessions and conversions.

Paid suites such as Semrush and Ahrefs add backlink data, competitor gaps, and scheduled site audits. They speed things up on bigger sites. They don’t replace GSC, because only Google can tell you what Google indexed.

How To Run An SEO Audit In 9 Steps

Step 1: Set The Goal And Get Read Only Access

Write down what triggered the audit before you open a single tool: a traffic drop, a redesign, a plateau, a new agency. A drop after a site migration points you straight at redirects and indexing. A slow plateau points at content and authority. The trigger decides where you dig first.

Then collect access: Search Console, GA4, the CMS, and ideally the hosting or CDN dashboard. If you’re bringing in an outside auditor, follow Google’s own advice: grant read access to Search Console at this stage, not write access.

Step 2: Check Whether Google Can Crawl Your Site

Open yourdomain.com/robots.txt and read it line by line. You’re looking for a Disallow rule that blocks folders you want ranked. A leftover “Disallow: /” from a staging site is rare, but when it happens it explains everything.

Next, check your XML sitemap. It should list only the canonical, indexable, 200 status URLs you want in search. Sitemaps full of redirects, tag archives, and noindexed pages send mixed signals and make GSC reports harder to read.

Finally, type the four versions of your domain into a browser: http, https, with www, without www. Three of them should redirect to the fourth in a single hop. If two versions load separately, you have a sitewide duplicate problem.

Step 3: Find Out What Google Actually Indexed

This is the step most DIY audits skip, and it’s often where the answer is. Go to Search Console > Indexing > Pages. The report splits your URLs into indexed and not indexed, with a reason for each. The reasons that deserve attention:

GSC status What it usually means
URL marked “noindex” A directive told Google not to index the page. Fine for thank you pages. A disaster on a service page.
Crawled – currently not indexed Google fetched the page and decided not to index it (for now). Often a quality or duplication signal. Google says there’s no need to resubmit.
Discovered – currently not indexed Google knows the URL but hasn’t crawled it yet, often because it expected crawling to overload the site. Common on large or slow sites.
Duplicate without user selected canonical Google considers the page a duplicate and picked a different URL as canonical, because you didn’t specify one.
Blocked by robots.txt Your robots.txt stopped Google from crawling it. Check this against your list from step 2.

 

Definitions paraphrased from Google’s Page indexing report documentation. One line from that page is worth remembering: Google says not to expect every URL on your site to be indexed. The goal is that every page you care about is indexed, not a 100% score.

For individual pages, use the URL Inspection tool in GSC rather than a site: search. Google itself notes the site: operator doesn’t necessarily return every indexed URL.

Step 4: Crawl The Site Like A Search Engine Would

Run Screaming Frog (or your suite’s site audit) across the whole site and sort the results by what they affect:

  • Status codes: 4xx errors on pages that still get internal links, and 5xx errors of any kind.
  • Redirects: chains of more than one hop, and internal links pointing at redirected URLs instead of the final one.
  • Canonicals: pages that canonicalize to a different URL by accident, often on paginated or filtered pages.
  • Crawl depth: important pages more than three or four clicks from the homepage get crawled less and signal less importance.
  • Orphan pages: URLs in your sitemap or GA4 that no internal link points to. Connect Screaming Frog to GSC and GA4 to find them.
  • Titles and H1s: missing, duplicated, or so generic that two pages describe themselves identically.

Ignore the urge to fix everything the crawler flags. Note how many pages each issue affects and which of those pages matter commercially. You’ll need both numbers in step 9.

Step 5: Check Page Experience With Field Data

Go to Search Console > Experience > Core Web Vitals. This shows field data from real Chrome users, grouped by similar URLs, which is more useful than a single lab score from PageSpeed Insights. Google’s thresholds for a good experience are LCP within 2.5 seconds, INP under 200 milliseconds, and CLS under 0.1.

Keep this in proportion. Google says good Core Web Vitals align with what its ranking systems reward, but passing them won’t lift a page that doesn’t answer the query. Fix the failing URL groups that contain your money pages, and fix them at the template level, since one slow hero slider on a shared template can fail hundreds of URLs at once. While you’re here, confirm the site is fully on HTTPS and that mobile pages show the same content as desktop.

Step 6: Audit Content Against Search Intent

Technical fixes get a page considered. Content decides whether it wins. Start with Search Console > Performance > Search results, filtered to your most important pages, and compare the queries they actually get impressions for against what the page was written to rank for.

Three problems show up again and again:

Intent mismatch. A page written as a sales pitch trying to rank for a “how to” query, or the reverse. Search the target query yourself and look at what ranks. If the top results are guides and your page is a service page, the page type is the problem, not the keywords.

Cannibalization. In the Performance report, click a query and open the Pages tab. If two of your URLs swap impressions for the same query, they’re competing. Merge them, or make their jobs clearly different.

Thin or commodity pages. Run key pages through the self assessment questions in Google’s helpful content guidance, such as whether the content provides original information or analysis, and whether it presents information in a way that makes you want to trust it. City pages that swap one place name and nothing else usually fail here.

While you’re on each page, check its structured data in the Rich Results Test. Valid Organization, LocalBusiness, Product, or Article markup helps pages qualify for rich results. It isn’t required for AI features, and it won’t fix a weak page.

If this is where your audit is turning up the most problems, that’s normal, and it’s also the step that takes the most time to fix. It’s the work we do inside our SEO services, if you’d rather hand it off.

Step 7: Check Backlinks And Local Signals

In Search Console > Links, look at your top linking sites and your most linked pages. Then compare your referring domains against two or three competitors who outrank you, using Ahrefs or Semrush. You’re looking for a gap in relevant sites linking to competitors but not to you, not a raw count.

Don’t panic over “toxic” link scores. Google says most sites will not need the disavow tool, and it should be reserved for a considerable number of spammy or artificial links that caused, or are likely to cause, a manual action. Check Security & Manual actions > Manual actions in GSC. If it says no issues detected, a disavow file is almost never your priority.

If you serve customers locally, audit your Google Business Profile here too: primary category, service areas, hours, and whether your name, address, and phone match your website and major directories.

Step 8: Check Your AI Search Visibility

In 2026 an audit that ignores AI answers is incomplete, but this is also where the most bad advice lives. Google’s guide to optimizing for generative AI features, published in May 2026, says its AI Overviews and AI Mode are rooted in core Search ranking systems. For Google, the steps above are the AI audit. The same guide says you can ignore llms.txt files, “chunking” content into tiny pieces, and special AI markup for Google Search.

What to check instead:

  1. Open the Generative AI performance report in Search Console to see how your pages appear in Google’s AI features, and confirm your site is included in generative AI features there.
  2. Check robots.txt and your CDN or firewall settings for rules that block AI crawlers such as OAI SearchBot or PerplexityBot, if you want to appear in ChatGPT and Perplexity answers.
  3. Ask the assistants your buyers use 10 to 20 questions they’d ask before hiring someone like you, without your brand name. Record who gets mentioned and which pages get cited.

ChatGPT and Perplexity aren’t bound by Google’s guidance, so treat that third check as its own baseline and rerun it on a schedule.

Step 9: Turn Findings Into A Prioritized Fix List

This is the deliverable. Everything before it is research. For each finding, record four things: the stage it belongs to (from the table near the top), how many pages it affects, whether those pages drive revenue, and how much effort the fix takes. Then sort.

Priority Stage Pages affected Effort
Fix now Find or Index Any money page Any
Fix this month Render, Understand Many pages or a key template Low to medium
Schedule Understand, Trust A handful of pages Medium to high
Park Any Low value pages only Any

 

A finding that blocks indexing on your main service page is a “fix now” even if the tool rated it a notice. A thousand missing alt tags on blog images is a “park” even if the tool painted it red.

What An SEO Audit Looks Like In Practice

Here’s a hypothetical to show how the order changes the outcome. The business and numbers are illustrative, not a client result.

Say an HVAC company runs a 60 page WordPress site. A site audit tool reports 212 issues: 150 missing meta descriptions, 38 images without alt text, 14 pages with low word count, and 10 other warnings. Search Console tells a different story. Of the 60 URLs, 22 aren’t indexed. Four carry a noindex tag, including the main “AC Repair” service page, left over from the redesign. Six city pages sit in “Crawled – currently not indexed” because each one is the same paragraph with a new town name. Nine are parameter duplicates from ad tracking links. On mobile, one template group fails LCP at 3.4 seconds because of a homepage style slider reused on every service page.

The fix list in order: remove the noindex from the four pages (minutes of work, and it unblocks the page that should earn the most leads). Rewrite or merge the six city pages with real local detail. Swap the slider on the shared template. Add canonicals and clean the sitemap. The 150 meta descriptions go last, on the pages that matter. The biggest problem on the site was one line in a 212 line export.

How To Tell A Real SEO Audit From A Tool Export

If someone hands you an audit, whether it’s free or a paid project, run it past these questions:

  • Does it use your Search Console data, or only a third party crawler?
  • Is every finding tied to specific URLs and to business impact, not just a severity color?
  • Is there a ranked fix list with an estimate of the work involved? Google’s own guidance says an evaluation should give realistic estimates of improvement and of the effort needed.
  • Does it say what not to fix?
  • Does it promise rankings? Google is blunt here: no one can guarantee a #1 ranking, and you should be wary of anyone who claims to.

An audit that fails the first three is a crawl report with a cover page.

How Often Should You Run An SEO Audit?

Most guides suggest a full audit once or twice a year with lighter checks quarterly, and that’s a sensible default. We’d tie it to events instead of the calendar: always audit before and after a migration, redesign, CMS change, or domain move, and whenever organic traffic drops in a way you can’t explain. In between, a 15 minute monthly look at the GSC Page indexing and Core Web Vitals reports catches most problems before they turn into a traffic drop.

Frequently Asked Questions

Q1: Can You Run An SEO Audit Yourself?

Yes, especially on a site under a few hundred pages. Search Console, GA4, and the free version of Screaming Frog cover most of the steps above. The hard part is prioritizing, and that’s where a second opinion earns its keep.

Q2: How Long Does An SEO Audit Take?

It depends on site size and how much data you have. A small site can be checked in a day. Large sites take longer, mostly because findings have to be sampled by template and validated against logs and analytics.

Q3: What Is The Difference Between An SEO Audit And A Technical SEO Audit?

A technical SEO audit covers steps 2 to 5: crawling, indexing, rendering, and speed. A full SEO audit adds content, links, local signals, and AI visibility, then ranks everything together.

Q4: Do SEO Audits Include AI Search Now?

They should. For Google, AI visibility mostly follows from normal SEO health, which Google confirmed in May 2026. For ChatGPT and Perplexity, add crawler access checks and a prompt based baseline.

Q5: What Should An SEO Audit Include?

A complete SEO audit checks crawlability, indexing, site structure, page speed and Core Web Vitals, content quality and search intent, backlinks, local signals, and AI search visibility. It should end with a prioritized fix list, not just a list of warnings.

Q6: Is A Free SEO Audit Tool Enough?

Free tools are a good starting point, but they only report issues. They can’t tell which problems affect your money pages or which warnings are safe to ignore. Use them to collect data, then apply judgment to decide what to fix first.

Recently Blogs

Free SEO Audit – Instantly Boost Your Visibility

“Our SEO has always worked hard to make sure that we’re getting what we need out of the partnership, not just what may seem like the best result”

Related Blog Posts