SEO proof report: did the change work, measured against pages you did not touch.
Two Search Console exports and the pages you changed. Get a verdict against pages nobody touched.
One-off $39. Back within 8 hours. No calls, no account access.
Performance on Search results
Total clicks
2,184
Total impressions
61.3K
Average CTR
3.6%
Average position
9.8
Verdict
+68 clicks a month
+71% against the control. 95% range 21% to 143%.
Three files, back within 8 hours.
The report
Every figure with the working behind it, so you or your agency can check it.
The verdict
The severance pay guide rewrite worked: about 68 more clicks a month.
+68
clicks a month
+71%
against the control
21–143%
95% range
Clicks a week, 8 weeks before the rewrite
Clicks a week, 4 weeks after
Search Console exports for the guide and the control, 4 August to 2 November 2026. A core update on 12 October moved both.
The action plan
Every fix written as a task, with the evidence and the check.
A one-page client write-up, under your name if you want
Ready to use as it comes.
The severance pay guide rewrite worked.
Clicks rose by about 68 a month more than on 19 guides nobody touched, over the same weeks. Your agency name goes at the top.
- A verdict, one of three
- It worked, it did nothing detectable, or it is too early to tell.
- The control group
- Untouched pages that tracked yours before the change.
- A placebo check
- The same test on untouched pages, to show the noise.
Then hand the plan to Claude.
Each task says what to change, where, and how to check it worked. Paste the file into Claude or Claude Code and ask for one task at a time, or send it to your agency.
Rewrite the guide I am pasting the way task 1 of this action plan describes. Keep every fact and figure.
Here is the notice period guide, rewritten the way task 1 describes. Every figure is unchanged:
Reply to Claude…
Did the SEO my agency billed for move anything?
The report answers each of these from your own data, not from your agency's monthly report.
The pages they changed, against pages nobody touched.
Changed pages +71% against untouched onesWhether a good season or a Google update is getting the credit.
A core update in the window, absorbed by the controlHow much movement is normal on your site, before anyone claims a win.
Normal weekly swing on this site: 18%
What you send.
About ten minutes of exports. The upload page walks you through each one after checkout.
Search Console export, beforeHow to get it
Search Console, Performance, a date range that ends before the change went live, Export, Download CSV.
Search Console export, afterHow to get it
The same property, a range that starts on or after the change. Two weeks or more is better, because Google has to recrawl first.
The pages you changed, and the dateHow to get it
Paste the URLs or a folder such as /blog/, and the day it went live.
SEO proof report, $39.
Back within 8 hours. If the report is not useful, reply to it and we refund you in full. Terms
Order for $39Questions
How do you prove an SEO change worked?
Against a control, not a before and after. A before and after credits the change with everything else in the window: seasonality, a core update, a good month for the industry. Matching the changed pages to untouched pages that tracked them before the change, then measuring the gap after it, removes most of that.
Can I do this myself for free?
Yes. The free Did it work? tool runs the same comparison in your browser. The report adds a person checking the result, the placebo check explained, and a write-up a client can read.
What does the SEO proof report not cover?
Whole-site changes and brand-new sites. A control needs pages nobody touched. Revenue, unless you send the leads the pages produced.
What do I do with the action plan?
Paste it into Claude, or open it in Claude Code next to your site's files, and ask for one task at a time. Each task names the change, where it goes, the evidence and how to check it worked, so you can check Claude's draft against the evidence. Or send the file to your agency as the list to work through.
How fast is it?
Within 8 hours of your upload. A person reads every report before it is sent. If something in your files stops the analysis, such as a missing column, we email you within the same window to say what to re-export.
Do I have to give you access to my accounts?
No. You download the files yourself and upload them after paying. Nothing connects to your accounts and nothing in them is changed.
What if it is not useful?
If the report is not useful, reply to it and we refund you in full. The money comes back to the card you paid with.
More on the method: how to prove an SEO change worked and how we measure. Or run the free Did it work? tool in your browser.
Other audits