Search Console Analyzer

Twenty analyses on your own Search Console export, concentration, brand split, your own position-to-CTR curve, striking distance, query clustering, traffic forecasting, volatility, weekday seasonality and a per-page diagnosis. Ends with the single highest-value thing to do next.

How to use it

  1. In Search Console, open Performance, set the date range as wide as it will go, and click Export → Download CSV. You get a zip containing Queries, Pages, Countries, Devices, Search appearance and Dates.
  2. Unzip it and drop the whole folder onto this page. Every file is parsed in your browser. There is no upload step, and you can disconnect from the internet before dropping the files if you would rather verify that than trust it.
  3. Read the verdict at the top first. It names the single highest-value thing to do next, which is usually not the thing the dashboard draws your eye to.
  4. Work down through the sections. Export the whole thing as Markdown, or as a prompt you can paste into an AI agent along with your own context.

What the numbers mean

Concentration
What share of your clicks comes from your top 1, 3, 5 and 10 pages. In the reference dataset, 5 pages out of 160 produced 76.1% of all traffic and 96 pages earned fewer than ten clicks in seven months. If your top five is under about half, you have genuine breadth; over 80% and you have a dependency rather than a channel.
Brand vs non-brand split
Queries containing your name, separated from everything else. In the reference dataset this was 42.3% of all clicks at 16.02% click-through, against 3.29% for non-brand. Brand traffic is your most valuable and your least scalable: if it is over half your total, you are being remembered rather than discovered.
Your own position→CTR curve
Every published CTR-by-position table is an average across categories that behave nothing like each other. This measures yours from your own rows, which is the only version that can price a ranking change for you.
Striking distance
Queries sitting roughly between position 5 and 15 with real impression volume behind them. These are the cheapest wins available, because the page already exists and already ranks. It is a title, an intro and an intent-match problem rather than a build.
Wasted visibility
High impressions producing almost no clicks. In the reference dataset one cluster returned 47,025 impressions and 102 clicks, a 0.22% click-through rate at average position 8.42. On a dashboard that highlights impressions this looks like success. It is essentially nothing, and it is the most commonly misread pattern in Search Console.
Volatility
How large a day-to-day swing is normal for you, so you can tell noise from a change. In the reference dataset the day-to-day range was 187% of the mean, traffic could halve and double back inside a week with nothing having been touched.

Questions

Is my Search Console data uploaded anywhere?

No. Every file is parsed and analysed in your browser using JavaScript. There is no server, no account and no storage. You can load this page, disconnect from the internet, and the tool still works, which is the only version of that claim you should accept from anyone.

Which files does it need?

It reads the standard Search Console CSV export. Queries and Pages give you the most, but it will also use Countries, Devices, Search appearance and Dates if they are present. Drop the whole unzipped folder and it works out what it has.

Why is my click-through rate falling while clicks go up?

Because you are ranking for more queries than before, and the new ones sit lower. In the reference dataset CTR fell from 29.25% to 2.72% while clicks rose 529×. Both numbers are true at once, and the falling one is usually the sign that the growing one is working.

How long a date range should I export?

As long as you can get, and at minimum four months. Search results settle on a one-to-four month delay, so a 28-day window is a report on work you have already forgotten doing. The reference dataset covers 224 days.

Can I run this on a client site?

Yes. Nothing leaves the browser, which is usually the blocker with client data. If you need an offline copy, the toolkit includes the same analysis as a dependency-free Node script generated from the same source code as this page.

Guides that use this tool

Other free tools