use casesbeginner10 minutesOctober 15, 2026

How to measure whether an SEO change actually worked

Compare the changed page's Search Console clicks, impressions, CTR and position over equal periods before and after the change, then check organic conversions in GA4. Here is the method, how long to wait, and how Cogny closes each change as worked, partial, didn't work or inconclusive.

How do I measure whether an SEO change actually worked?

Write down the page, the change and the date, then compare Search Console clicks, impressions, CTR and position for that page before and after, over equal periods. Check GA4 for conversions from organic. Cogny does this for every content and SEO change you ship through it: it reads your own Search Console and GA4 data seven days later and closes the change as worked, partial, didn't work or inconclusive.

How long should I wait before judging an SEO change?

Wait at least until search engines have recrawled the page and you have a full week of data after the change, so every weekday is counted once. Compare it with the week before the change, not with a different season. A small page may need several weeks before the numbers mean anything. If the data is still thin after that, the honest answer is inconclusive, not a win.

What counts as worked, partial or inconclusive?

Decide before you look. Worked means the metric you aimed at, say clicks or CTR on the changed page, moved clearly in the right direction and nothing important fell. Partial means it moved, but less than expected, or one metric rose while another dropped. Didn't work means no change or a fall. Inconclusive means too little data, or something else changed at the same time, such as a core update or a migration.

Can my AI do this automatically?

Yes. Connect Search Console and GA4 to Cogny, then use Claude, ChatGPT, Cursor, Codex or the Cogny app. When a change is shipped through Cogny, it becomes a ticket. Seven days later Cogny reads the page's own Search Console and GA4 data, compares the periods and closes the ticket with a verdict. Misses stay on the record. This runs on every plan, Free included, even at zero credits.

How to measure an SEO change, step by step

Log the change. Page URL, what changed (title, meta description, H1, internal links, new section), the date, and what you expect to move.
Pick one primary metric. Clicks for traffic work, CTR for title and description changes, average position for content depth.
Pull the before period. In Search Console, filter Performance by the exact page and take the 7 or 28 days before the change.
Wait, then pull the after period. The same length, starting the day after the change shipped.
Check the queries. Compare the page's top queries in both periods. A gain on the query you targeted counts more than a gain on an unrelated one.
Check conversions. In GA4, look at organic sessions and key events landing on that page for the same two periods.
Write the verdict. Worked, partial, didn't work or inconclusive, with the numbers next to it. Keep the misses; they tell you what not to repeat.

Frequently asked questions

Should I compare against the same period last year?

Use it as a sanity check for seasonal pages, but the main comparison should be the period right before the change. Last year's page often had different content, links and rankings.

What if I changed several pages at once?

Measure each page on its own, and add up the group. If the whole site moved at the same time, a sitewide cause such as an algorithm update is more likely than your change.

Does Cogny split traffic like an A/B test?

No. Cogny reads before and after in your own data. It does not split traffic, so a verdict is a before-and-after reading, not a controlled experiment.

Which plans measure changes?

Every plan, Free included. Measurement runs even when a workspace has no credits left.