Free Tool
Plan one SEO change, record what Google Search Console says before and after, and get an honest read on whether it worked. No API keys, no Apps Script, no subscription. Just a Google Sheet you copy and fill in.
Free
no sign-up needed
90 Days
like for like before and after
5 Checkpoints
from day 7 to day 90
4 Numbers
per URL, per checkpoint
Why I Built It
Someone rewrites a page title. Someone adds a paragraph of copy. Someone tidies up an internal link. Six weeks later traffic looks a bit different and everybody agrees the change must have helped, or must have hurt, depending on who is in the room.
That is not measurement. That is memory, and memory is generous to whoever is telling the story.
The problem is rarely a lack of data. Google Search Console already holds the numbers. The problem is that nobody writes down what they changed, when they changed it, and what they expected to happen, so there is nothing solid to compare against later.
This template fixes that part. It gives one SEO change a start date, a hypothesis, a fixed set of checkpoints and a before and after comparison that is calculated the same way every time.
You record the change you made and the date it went live. The sheet works out the exact date ranges to export from Search Console, then does the sums for you.
Inside the Sheet
Ten tabs, and you only ever type into four kinds of them. Everything else calculates itself.
A short set of steps at the front of the workbook, so the sheet explains itself without you coming back here.
The test name, the implementation date, what changed, your hypothesis, the primary metric and the URLs being tested.
The 90 full days before your change went live, which everything afterwards is measured against.
Day 7, 14, 28, 56 and 90, each with the same four columns, so every entry is made the same way.
Normalised daily comparisons, every change calculated for you, and a breakdown of how each tested URL performed.
Headline numbers, the trend for your chosen primary metric, and a place to record the conclusion and what happens next.
How It Works
You need a Google account and Search Console access to the site you are testing. That is the whole list.
01
One copy per test, so each workbook holds one clean experiment rather than a pile of half-remembered ones.
02
Name the test, add the date the change went live, describe what you changed and what you expect to happen, choose your primary metric and paste in your URLs.
03
The sheet tells you the exact Search Console date range for every checkpoint and the earliest date you should export it.
04
In Search Console, go to Performance, then Search results, then the Pages tab. Enter clicks, impressions, CTR and average position for each URL, keeping your search type and filters identical every time.
05
The Results tab holds the normalised comparisons and the URL level detail, so you can see which pages moved and which did not.
06
Positive, negative, inconclusive or needs more time, then the decision that follows from it. A test with no recorded conclusion is a test you will run again by accident next year.
The Method
Two things quietly ruin most before and after SEO reporting. The sheet handles both, and it is worth knowing how before you trust the output.
Periods of different lengths. A 7 day window and a 90 day window produce wildly different totals even when nothing has changed, so clicks and impressions are converted to daily averages before anything is compared.
CTR and position. CTR is total clicks divided by total impressions rather than an average of averages, and average position is weighted by impressions, so a URL with nine impressions cannot shout as loudly as one with nine thousand.
The Search Console delay. Data arrives a couple of days late, and exporting a range too early quietly understates your most recent numbers. Every export date carries a three day buffer for that reason.
The direction of the numbers. Search Console counts position downwards, where 1 beats 11, which makes raw deltas read backwards. The sheet flips the sign, so a positive position change always means the ranking improved.
I would rather say this on the page than in an email afterwards.
This is a before and after measurement, not a controlled split test. There is no holdout group, so the sheet cannot prove your change caused the result. A core update, a competitor rebuild, a seasonal swing or a change somebody else made on the same site can all move the numbers underneath you.
What it can do is stop you fooling yourself. It fixes the dates before you see the outcome, it calculates both sides identically, and it returns a verdict of inconclusive when the result is inconclusive. A good number of SEO tests land there, and recording that honestly is worth more than a confident story built on a fortnight of data.
Day 90 is the one to judge the test on. It is the like for like comparison, 90 full days against 90 full days. The earlier checkpoints show the direction of travel, not the answer.
Rather Not Run It Yourself?
Running the test is the easy half. Deciding what to change first, and what to do with a result once you have one, is where most of the value sits. That is the part I do for clients.
Tell me about your site and I will give you a straight answer on the change most likely to be worth measuring, whether that is titles, structure or something underneath the bonnet.
Questions
Something else on your mind? Ring or WhatsApp me. You will get a real answer from me, not a sales script.
Making a deliberate change to a page or a set of pages, then comparing search performance before and after over a fixed period to see whether the change helped. The point is deciding how you will measure it before you see the result, rather than hunting for a favourable comparison afterwards.
No. Search Console and a Google account are enough. There is no API key, no Apps Script, no plugin and nothing to pay for.
As many as you are prepared to type four numbers for at each checkpoint. The list you enter in Test Setup flows through to every checkpoint tab and into the results automatically. In practice, a focused group of pages that all received the same change gives a much cleaner read than a long mixed list.
It is long enough for Google to recrawl, reassess and settle, and it covers a full quarter, which smooths out monthly seasonality. Shorter windows are where most premature conclusions come from. The earlier checkpoints are there so you are not sitting in the dark for three months.
You can, but the result will not tell you which change did the work. One change per copy of the sheet.
Record it as inconclusive and move on to the next test. A null result is still information, and it is a good deal cheaper than rolling the same change out across the rest of the site on the strength of a hunch.
Yes. Make a copy and it is yours. There is no email form in front of it and no trial that expires. You can find my other free marketing tools here too.
It is built for Google Sheets and that is where I have tested it. The formulas may not survive an export to Excel, so I would stick with Sheets.