How to Do an SEO Audit: A Step-by-Step Process

There are two ways people mean "how to do an SEO audit." One is "give me the list of things to check" — that's the SEO audit checklist, and you should keep it open in another tab. The other is "walk me through the actual process on a real site," which is this. A good audit isn't ticking boxes in order; it's a sequence where each step tells you where to look next.

Here's the process I run, in the order I run it, with the tools for each step and a note on what's worth your time versus what isn't.

Before you start: pick a scope and a snapshot

Decide whether you're auditing the whole site or one section, and grab a baseline first — current traffic, rankings, and indexed page count — so you can tell later whether your changes worked. An audit with no "before" is just opinions.

Budget a few hours for a small site, a day or more for a large one. You don't need to finish in one sitting; you need to finish before you start changing things.

Step 1 — Crawl the site and build your inventory

Run a crawler (Screaming Frog's free tier handles small sites; there are hosted options for big ones) to get every URL, status code, title, and meta in one export. This is your map. The pages you forgot existed — old campaign URLs, duplicate tag pages, a staging path that got indexed — are usually where the problems hide.

What you're looking for: 404s, redirect chains, duplicate titles, missing metadata, and anything indexed that shouldn't be.

Step 2 — Check indexation and technical health

Open Google Search Console and go to the Pages (Coverage) report. This tells you what Google can actually index and why it's excluding the rest. A page that isn't indexed can't rank — so this is where you catch the expensive problems: noindex tags left on by accident, robots.txt blocking a whole section, canonical tags pointing at the wrong place.

While you're here, check Core Web Vitals and mobile usability. You don't need a perfect score; you need to not be in the red.

Step 3 — Audit on-page and site structure

Now go from "can Google reach it" to "is each page set up to be understood." Spot-check your most important pages for clean title tags, a single clear <h1>, sensible URLs, and — the one people skip — internal linking. Orphan pages (no internal links pointing to them) are invisible to both crawlers and readers. Make sure your important pages are linked to, not just from.

This is also where you catch keyword cannibalization: two or three pages competing for the same query and splitting the result so none of them wins. Consolidate them.

Step 4 — Audit content quality

This is the step most audits rush, and on most sites it's where the biggest gap lives. Steps 1–3 tell you whether Google can rank your pages. This step asks whether they deserve to.

For each page that matters, read it like a skeptical stranger and ask: does it answer what the searcher actually wanted? Are the facts right and sourced? Does it show real experience, or is it a confident summary of things someone Googled? Is it original, or a reshuffle of what already ranks? Google's systems increasingly grade content on exactly these lines — "helpful, reliable, people-first" content and E-E-A-T — and a technically perfect page that fails here will still go nowhere.

The catch is that you can't grade your own content objectively; you wrote it, so it reads fine to you. That's the specific job we built Revylo for — it scores a URL against those quality dimensions and returns what's holding it back, per page. (We ran it on our own sites first and published the results, bad scores included, and the methodology is public if you want to see how the number is built.) For more real audits — what failed, why, and what we changed — see SEO content audit case studies. If you'd rather assess it by hand, the content-audit guide covers the manual version.

Step 5 — Review the backlink profile

Pull your backlinks (Search Console's Links report is free; dedicated tools go deeper). You're looking for three things: links worth keeping and reinforcing, lost links worth reclaiming, and genuinely toxic links worth disavowing. Don't disavow aggressively — most "toxic" links are harmless, and over-disavowing does more damage than the links did.

Step 6 — Verify your tracking before you trust any of it

Put this near the end and treat it as a gate: confirm Search Console and GA4 are actually firing, that traffic is attributed correctly, and that your conversion events fire. If your analytics is wrong, every conclusion you drew above is built on bad numbers. (Plenty of "traffic" turns out to be bots or mis-tagged campaign clicks that never represented a real visitor — confirm the plumbing before you act on the dashboard.)

Step 7 — Prioritize and build the action plan

You'll finish with a long list of issues. Don't fix them in the order you found them. Rank by leverage:

Anything blocking crawling or indexing (Step 2) — highest priority, always.
Tracking integrity (Step 6) — so you can measure the rest.
Content quality (Step 4) — usually the biggest mover.
On-page, structure, and links — the finishing work.

Turn the top handful into concrete tasks with owners and dates. An audit that ends in a 60-row spreadsheet nobody acts on was a waste of a day.

Step 8 — Re-audit on a cadence

SEO isn't static — facts go stale, competitors improve, Google updates its systems. Re-run the audit quarterly on an active site so you catch decay before it compounds. The fastest way to keep the heaviest step current is to re-score your key pages' content quality regularly rather than waiting for a traffic drop to tell you something broke.

That's the whole process: crawl, index, on-page, content, links, tracking, prioritize, repeat. Keep the checklist open as you go, and when you hit Step 4, run your key pages through Revylo free — it's the one step a checklist can't do for you.