An unnecessarily important SEO experiment

How long does it take Mark Williams-Cook to cook an omelette?

The page began with a deliberately ridiculous zero-volume query. It became a small test of how an AI Overview citation and a standard blue link appeared in Search Console.

The short answer

The original culinary answer was approximately 15 minutes, or as little as five minutes when the ingredients and kitchen were ready.

The SEO answer became more entertaining: according to the experiment, it took approximately one Search Console impression to cook the omelette.

Why did this page exist?

I wanted a query with no meaningful existing search demand so that manual checks would be easier to recognise in Search Console. The question about Mark Williams-Cook and an omelette was specific, harmless and unlikely to be searched accidentally.

The page was indexed and shown as both a supporting citation in an AI Overview and a standard organic result. Mark Williams-Cook then helped by reposting the experiment.

The method

  1. Publish and index a page targeting the zero-volume omelette query.
  2. Prompt the search experience so the page appeared as an AI Overview citation.
  3. Confirm that the same URL also appeared as a standard organic blue link.
  4. Perform two manual checks at different times.
  5. Review the exact query and page in Search Console after the data had time to appear.

What appeared in Search Console?

The test produced two impressions, matching the two manual result-page checks. It did not show an additional impression simply because the same URL was also cited inside the AI Overview.

That was a tiny controlled observation, not a universal law. Search interfaces and reporting change, and Google began testing dedicated generative-AI visibility reports in Search Console later in 2026.

The useful lesson was not the size of the number. It was that a clearly isolated query made the reporting behaviour easier to inspect.

What can an SEO learn from an omelette?

  • Design tests around one question instead of interpreting a noisy site-wide total.
  • Record the URL, query, time and exact action taken.
  • Distinguish a citation, an impression, a click and a conversion.
  • Avoid turning a two-impression observation into a sweeping claim.
  • Use memorable experiments; people are more likely to discuss and challenge them.

And the actual omelette?

Preparation explains most of the difference between five and 15 minutes. Ingredients that are ready, a heated pan and familiar tools make the faster version possible. The same principle applies to technical work: preparation often looks like speed after the fact.

If you came here for a recipe, the original sources remain the best place to settle the culinary details. If you came here for SEO, the experiment is documented in Tony Ruiz's LinkedIn post about the test.

Editorial note: The displayed spelling and grammar have been corrected, and the 2026 experiment context has been restored. The exact URL remains unchanged so existing Search Console history and links are preserved.

Have a less culinary measurement question?

Traffic Roosters can help separate what the available data confirms from what still needs testing.

Ask Traffic Roosters