SEO test results analyzer

Verdict with confidence level for a finished SEO split test

This tool is planned. Sign up and we’ll email you when it ships.

Planned cost: 1 credit

PlannedSEO test results analyzer
Input
your topic
Options
Default
Runuses 1 credit
Output
First result
Second result
Third result
Illustration — sample content, not a screenshot or a real resultNotify me when it’s live

Overview

What it does

Analyses a finished SEO test and says whether the change you saw is meaningful.

Method

How it works

1. Upload before and after

Search Console exports covering the period before the change and the test period, plus your test and control URLs. A saved plan from the split test planner fills this in for you.

2. It compares both groups

Not test-before against test-after — that measures seasonality. It compares the change in the test group against the change in the control group over the same weeks.

3. You get a verdict and a confidence level

Plain English, with the number behind it. “No detectable difference” is a real and common result, and the tool says so rather than finding something.

4. You decide what it means

The tool reports the statistics. Whether a 4% lift justifies the work is your call.

Example

Sample output

Illustrative example. The tool is not connected yet, so this shows the shape of the output rather than a real run.

Input

Six weeks of data, 71 test pages and 71 control pages.

Output

Verdict: The test group's click rate rose relative to control. The difference is unlikely to be chance.

  • Test group: click rate up

  • Control group: roughly flat over the same period

  • Difference-in-differences: positive

  • Confidence: high

Caveat shown with every result: this measures what happened on these pages in this period. It is evidence, not a law.

FAQ

Common questions

What does a run cost?

What does a run cost?

One credit. The statistics run in code and only the plain-English interpretation calls a model, so it sits at the base rate. Free gives you one analysis at a time, shown on screen. Pro adds saved history and a shareable report, which matters if you are reporting results to someone else. See pricing. A shareable report is the main reason people move to Pro for this one, since a verdict nobody else can see is hard to act on.

Do I need to have used the planner first?

Do I need to have used the planner first?

No, but it helps. You can upload before-and-after exports with your own test and control URLs. If you planned the test with the split test planner, a saved plan fills in the groups for you and removes the most common source of error, which is a control group that was never comparable to begin with. Without a plan, check that your control pages were comparable before the test started, not just after.

What does the confidence level actually tell me?

What does the confidence level actually tell me?

How likely it is that the difference between your two groups is not chance. It does not tell you the change was worthwhile, or that it will hold, or that it would work on another site. High confidence in a tiny effect is still a tiny effect. The tool reports the statistics; whether the result justifies the work is your call. It reports the number rather than interpreting it for you, which is deliberate. See the split test planner for setting the test up so the number means something.

Does it ever say nothing happened?

Does it ever say nothing happened?

Often, and that is a real result. Most SEO changes do not move anything detectable, and a tool that always finds something is not measuring — it is reassuring you. "No detectable difference" saves you from rolling out a change that did nothing. The AI SEO tools guide covers why that matters when judging any tool's output. A flat result on a change you were sure about is information worth having.

Is this tool available now?

Is this tool available now?

Not yet. It’s on the SERPish roadmap. Join the waitlist and you’ll get an email when it’s live.

More tools

Related tools

SEO split test planner

Planned

Beta

Live

Title and meta writer

Write title tags and meta descriptions based on your keyword, page type, and what the page is actually about.

Title bet

Planned

Get new SERPish things, when they’re ready.

Subscribe

Get new SERPish things, when they’re ready.

Subscribe