Is my SEO agency doing a good job?

A monthly report written by the agency it grades

A monthly SEO report is an activity report: it shows what the agency did and what it measures, with its own tools. It can't tell you whether the site the agency maintains is still readable by search engines, page by page.

Several technical flaws go unnoticed when you browse the site. A page moved without a redirect returns an error to anyone following the old link, search engines included. A site that answers "page found" for a URL that doesn't exist lets search engines index empty pages.

And if Search Console is in the agency's name, the data you would judge it by sits with the agency.

Keep Search Console in your own name

Search Console is Google's tool for site owners. Clicks, impressions, the queries that bring visitors, average positions and the links pointing to the site all show up there. That is where an agency's results are judged, and no outside audit can see those numbers.

Owners can add and remove users, configure settings and view all the data, as Google's help page on owners, users and permissions explains. The agency can work there with full access without being an owner. The same page warns that a property must have at least one verified owner, or no users will have access to it: if that owner is the agency, your access depends on it.

If you part ways, removing the agency from the list isn't enough: as long as its verification tokens stay in place, it can verify ownership again.

Questions that call for answers you can check

Google's page "Do you need an SEO?" lists questions to ask before hiring one, including this one: what results do you expect to see, in what timeframe, and how do you measure success? It also states that no one can guarantee a #1 ranking on Google, and that an SEO with access to your server should be willing to explain all the changes made to your site.

Once the agency is working on your site, ask about the state of the site:

• Which pages should be in Google, and which are left out on purpose?
• When a URL changes, who sets up the redirect, and to which page?
• What structured data does the site declare, and what does it describe?
• What did you change on the site this month, and why?
• What does each exclusion reason in the Page indexing report mean for us?

Search Console's Page indexing report lists the URLs Google hasn't indexed, with a reason for each: page with redirect, soft 404, URL marked "noindex", duplicate without user-selected canonical. Google adds that you shouldn't expect every URL on your site to be indexed, so an excluded page is not a mistake in itself. What you can expect from the agency is a reason for every exclusion that affects a page you care about.

When pages are missing from Google, the guide to a website not showing on Google separates the causes visible from outside from the ones that need Search Console.

An outside review when the agency starts, and after each redesign

You check the agency's answers against a review it didn't run. That is what a technical SEO audit by a third party does: it tells you whether search engines can read the site the agency maintains, and where they can't. Rankings stay in Search Console.

Three moments suit it: when the agency starts, to set a baseline; after a redesign, a migration or a template change; and before you renew the contract. A baseline changes the conversation: at the second review, what the agency fixed shows, and so does anything that broke since.

Indexing, tags, redirects: the state the agency maintains

These percentages cover about 30 websites we audited between August 14, 2026 and September 24, 2026. Each one counts only the sites where the point could be checked. Many were audited because a defect showed up quickly, so these figures describe our audits, not websites in general.

This part of an agency's work shows on the site itself. The audit reads it on every page analyzed, and names the page involved for each flaw:

• Which pages Google can index: indexing directives, canonical URLs, the robots.txt file and the sitemap.
• What the search engine sees before any script runs. On the site's reference page, usually the home page, what the server sends is compared with what the browser renders, and a canonical URL or an indexing directive that exists on one side only gets flagged. The guide to SEO for JavaScript-rendered sites explains why that gap matters.
• The tags on each page: a title and a description that are present, the right length and different from page to page, and a single main heading. A tag placed outside the page head is ignored by search engines, and it gets flagged too.
• Redirects, followed from the http and https versions of the address, and dead links, found by crawling the site. A multi-hop redirect forms when a URL that already redirects gets redirected again.
• The response to a URL that doesn't exist. A site that answers "page found" instead of a 404 error lets search engines index empty pages; Search Console calls this a "soft 404". The guide to what a site should return for a missing page explains why.
• Structured data in the page code: whether it is there, whether it can be read, and the fields each type must carry.
• Language versions, when the site has more than one: each version has to declare the others, and those declarations have to match both ways.

Less often, the audit finds a canonical URL or an indexing directive that only exists after scripts run, or a site in several languages that doesn't declare its versions. None of these flaws is visible when you browse the site.

One store that ranks well, one business site to fix by template

A report says what holds up as well as what is missing. That is how you judge an agency's work without blaming it for what it got right. The two samples we publish show it in opposite situations:

• In the sample report on an online store, SEO is the site's strongest point: content comes straight from the server, every page has a title, a description and a canonical URL, and none of the internal links followed is broken. The caveats are specific. A quarter of the descriptions run longer than Google displays, and seven pages are set to "noindex": for five listing pages, the report calls it a defensible editorial choice; for a product on sale, it is a page search engines won't show. That is exactly the kind of exclusion the agency should be able to explain.
• In the published report on a B2B business website, the foundation is in place: sitemap, robots.txt, language versions declared both ways. Four template defects weigh on the whole site: no canonical URL even though several URL variants serve the same content, no structured data, a footer logo that links to a missing page, and four sales pages that no link reaches. Page by page, the report also finds three case studies titled with the brand name alone and no description, and an 87-character home page title that Google will cut off.

The second report sums it up as template fixes: few in number, broad in effect. That is a list an agency can take on, and a second audit then shows what was done. The criteria follow a published method the agency can read just as you can. Every limit of the review is stated in the report. A point we could not verify is never presented as compliant.

The first shows the other side: SEO at its best on a site where compliance is the weakest pillar. The audit covers all 9 pillars, not just SEO, and the guide to how an SEO audit differs from a website audit explains what each one looks at. Plans start at €200 excl. VAT and differ only in the number of pages analyzed; they are listed on the pricing page. Delivery takes 5 working days up to 60 pages, with payment collected on delivery.