Search Console reports that matter for clinics

Search Console is free, it reports on your own property using Google's own data, and most clinics use perhaps a tenth of it. This article covers what to look at, in what order, and what to stop reporting.

Performance: the query report

The Performance report shows impressions, clicks, click-through rate and average position, filtered by query, page, country, device and search appearance. The filtering is the whole value. The unfiltered totals are close to meaningless because they average across a property with many different jobs.

The four filtered views worth building.

1. Queries by page for your top ten pages. This tells you what each page is actually being shown for, which is frequently not what it was written for.
2. Pages with high impressions and low click-through. These are pages being shown and not chosen, which is usually a title and description problem, and it is the cheapest fix available.
3. Queries containing question words. These are the real questions, in patients' words, and they are content research nobody else has.
4. Branded versus non-branded, split by filtering on the clinic name. Growth in non-branded impressions is discovery growth. Growth in branded is demand created elsewhere.

Pages: the indexing report

The Pages report groups every known URL by whether it is indexed and, if not, why. Read it by status and fix the largest mechanical cause first. The statuses and their meanings are covered in detail in the indexing article on this site, and the discipline here is to treat the report as a work queue rather than as a score.

Watch the trend of non-indexed URLs after a content change. A rise following a bulk page creation is the clearest possible sign that the pages were too similar to each other.

URL Inspection: the diagnostic

URL Inspection reports on a single URL: whether it is indexed, when it was last crawled, which canonical Google selected, whether the page is mobile usable and what enhancements were detected. The live test shows the fetched HTML and a rendered screenshot.

Two habits are worth forming. Compare the crawled HTML with what you expect to be served, which catches rendering problems immediately. And check the Google-selected canonical against your own, which catches canonicalisation problems that are otherwise invisible.

Core Web Vitals: the field report

This report groups URLs by their real-world performance, which is the measurement assessment is based on. Groups are formed by similar pages, so a fix applied to a template moves a whole group. Allow several weeks for the trailing window to reflect a change before concluding anything.

What to stop looking at

• Property-level totals with no filter applied.
• Average position as a headline metric.
• Day-to-day movement on any metric, which is noise at clinic volumes.
• Enhancement reports for features you do not use.
• The links report as a measure of anything actionable. It is a description, not a lever.

Setting it up properly

1. Verify a domain property so every subdomain and protocol is covered in one place.
2. Submit the sitemap and check the read status rather than assuming it was accepted.
3. Add every person who needs access, so it survives staff changes.
4. Connect it to your analytics if that is useful, but keep the query data as the source of truth for search.
5. Export monthly, because the interface retains a limited history and the long series is where the value accumulates.

Point five is the one that gets skipped and the one people regret. A monthly export takes two minutes and produces a multi-year record the interface will not give you later. Google's setup documentation covers verification.

A monthly review that takes twenty minutes

1. Non-branded impressions, this month against the same month last year.
2. The five queries with the most impressions and their click-through rates.
3. Any page whose impressions fell by more than a third, with a reason established.
4. New non-indexed URLs since last month, with the cause identified.
5. Any Core Web Vitals group that changed status.
6. Three new question queries worth writing about.

That is the whole review. It answers what happened, why, and what to do next, which is more than most clinic reporting achieves in ten pages.

Using the report to rewrite titles

The highest-yield use of the Performance report on a clinic site is finding pages with impressions and poor click-through, then rewriting the title and description to match what people are actually searching for. The queries are in the report. The current title is on the page. The gap between them is usually obvious once both are on screen together.

Rewrite the title to state what the page answers rather than to state the clinic's name twice. Rewrite the description as a sentence a person would want to read, under the length that gets truncated. Then note the date, because you will want to compare click-through before and after and nobody remembers when the change was made.

Filters worth saving

• Queries containing your clinic name, to separate branded demand from discovery.
• Queries containing question words, as a running content research feed.
• A single page, by query, for each of your top ten pages.
• Mobile only, since that is where most clinic traffic actually is.
• The last 28 days against the same period a year earlier, which removes seasonality.

Save these as a routine rather than rebuilding them each month. The consistency is what makes the numbers comparable, and comparability is the only property that makes any of them worth reading.