SEO audit report sample, and the template behind it

Search for an SEO audit report sample and most of what comes back is sales collateral: a well-designed PDF with the cover, the contents page and the methodology intact, and the actual findings replaced with placeholder text. You can see what the report looks like and not what it says, which is the only part that matters.

This page is the opposite. The structure below is the one I use, the worked example is a real finding written out in full, and the numbers in it are from this site rather than from an imaginary client.

The seven parts

1. What was checked, and what was not

One short paragraph. Which property, which date range of Search Console data, whether server logs were available, and what was out of scope. This exists so that six months later, when something else breaks, anyone can tell whether this audit would have caught it.

2. The answer, in three sentences

What is wrong, what it is costing, what to do first. Written before the detail and read by everyone who will never reach page two.

Most reports open with methodology instead, which is the author telling you they did the work rather than telling you what they found.

3. Blocking faults

Anything preventing pages from existing in Google. Usually zero items, occasionally one item that is worth more than the rest of the document combined. It gets its own section rather than being sorted into a severity list, because the difference between “this page ranks badly” and “this page does not exist” is a difference in kind.

4. Findings that cost traffic

The body of the report. Each one written to the same shape, covered below. Ordered by what it costs, not by category and not by how easy it is to fix.

5. Findings that are true but cheap to ignore

A deliberately titled section, and the one clients tell me they find most useful. It contains the things a crawler flagged that I looked at and decided are not worth your money. Naming them does two things: it proves they were considered rather than missed, and it stops someone else selling you the same list next quarter.

6. The content pile

Separate from the faults, because it is a different budget and often a different decision-maker. Pages competing for the same query, pages with no reason to exist, topics with no page at all. Technical SEO vs content SEO covers why the split matters.

7. What to do, in order, with who does it

Not a table of severities. A sequence, with each item marked as mine, yours, or your developer’s, and with the check that proves it worked. A fix nobody verified is not a fix.

One finding, written out in full

This is the shape every finding in section 4 takes. Four parts, in this order, no exceptions.

Finding: 22 pages rank inside the top 30 and take almost no clicks.

Evidence. Search Console, 8 June to 5 September 2026. Twenty-two URLs sit at position 30 or better. Together they return roughly 1,700 impressions and twelve clicks. The worst single case is /blog/url-is-not-in-property/ at average position 12.6 with 183 impressions and zero clicks.

Consequence, and the honest size of it. Smaller than it looks. Only eight of the twenty-two are inside position 21, where a better snippet can plausibly change behaviour: 562 impressions returning 4 clicks. At a normal click-through rate for those positions that is about nine clicks, so the realistic upside for the whole exercise is five clicks a quarter. The other fourteen sit between 22 and 31, where click-through is near zero whatever the title says. They have a position problem, and no title rewrite touches a position problem.

Fix. Rewrite title and meta description on the eight pages inside position 21, query early and in the searcher’s own words. Leave the other fourteen alone. Half a day. Verify by comparing clicks and CTR for those eight URLs over the following 90 days against the same window.

Three things about that finding are the point of this whole page.

It names its own evidence with a date range, so it can be checked and so it expires honestly. It states the consequence as a number and then argues that number down, because the finding is worth less than it first appeared and the client is entitled to know that before spending on it. And it says which part of the work to skip.

That last one is the difference between a report and a list. Anyone can tell you twenty-two pages have weak snippets. The work is establishing that fourteen of them are not worth touching.

For the record, that finding is from an audit of this site, and it corrected an earlier conclusion of my own that had read a position problem as a click-through problem.

The template

Copy this. There is nothing proprietary in it.

1. Scope
   Property audited, data range, access available, explicitly out of scope

2. The answer
   What is wrong / what it costs / what to do first, in three sentences

3. Blocking faults
   [each: evidence, consequence, fix, verification]

4. Findings that cost traffic
   [each: evidence, consequence, fix, verification]
   ordered by cost, not by category

5. True but not worth fixing
   [each: what it is, why it is not worth your money]

6. Content findings
   Separate pile, separate budget

7. Sequence
   Ordered actions, each marked mine / yours / developer,
   each with the check that proves it worked

The structure takes ten minutes to adopt and does not make a report good on its own. What makes it good is refusing to include a finding you cannot state the consequence of, and cutting the list until every remaining item would survive someone asking why it is there.

What is not in my reports

A score out of 100. It is a weighted average of things that do not weigh the same. A site can score 94 with a blocking fault and 61 with nothing wrong.

Screenshots of tool dashboards. Evidence that a tool was run, not that anything was thought about.

Every 404 on the site. What matters is the ones linked to from your own pages and the ones on URLs that used to rank. The rest is a count, and counts generate busywork.

Recommendations with no owner. “Improve internal linking” is not an action. Someone has to be able to do it on a Tuesday.

Anything about crawl budget on a small site. It is a real constraint in the hundreds of thousands of URLs and a way to sound technical below that.

The checks that produce all of this are published in full in the technical SEO audit checklist, in the order I run them. What an audit costs, and why nobody quotes it blind, is covered in SEO audit cost.

The short version

A useful SEO audit report has seven parts, and the two that most reports lack are the section admitting which findings are not worth your money and the sequence saying who does what next. Every finding names its evidence with a date range, states the consequence honestly even when that means arguing it down, gives a fix, and gives the check that proves the fix worked. Length is not a quality signal, it is the opposite.

If you want one run against your site rather than a template to run it yourself, that is the audit, and the diagnosis conversation before the quote costs nothing.