Should you opt out of Google AI search?

What changed in August.

Google says its new AI-search control and visibility insights completed their worldwide rollout on August 31, 2026. The control lets a site owner decide whether their content and links participate in AI Overviews, AI Mode and AI Overviews in Discover.

Opting out removes the site's traffic and impressions from those features. Google says the choice is not a ranking signal for results outside them. That statement does not promise unchanged total traffic. Google's announcement explains the boundary.

A business owner can reasonably dislike having their work summarized elsewhere. Another may welcome a mention that helps someone discover the company. The control gives each a choice. It cannot calculate which choice is worth more to your business.

Start with the pages in the report.

The new Search Console reports show generative-AI visibility, including impressions, pages and countries, with device information in the Search report. These appearances also remain part of the broader performance reports. Google describes the reporting here.

Use that information to identify the pages worth investigating. A popular information page and a product page with available inventory have different jobs. Put the page's purpose beside its visibility before you judge the number.

The reported impressions do not establish enquiries, completed orders or profit. Keep those records alongside the search data. Our earlier note on AI citations and referral traffic covers that measurement distinction.

Give the customer a reason to continue.

Consider the fictional florist in the illustration. A search answer might describe where the shop delivers. Someone ordering flowers still needs to choose an arrangement, confirm availability for a date and check the total for their delivery address.

The website should make those jobs easy. A generic paragraph about beautiful flowers leaves the customer with more searching to do. Accurate stock, original photography, clear delivery terms and an order path give them something useful to complete.

That is a website improvement you can make before reaching a final AI-search decision. It also helps customers who arrive through an ordinary result, an advert or a recommendation. For service businesses, the equivalent may be seeing the actual work, checking the service area and understanding what happens after an enquiry.

These examples are recommendations about customer usefulness. They are not a formula for winning an AI citation.

Before changing participation.

1. Record the starting point.

   Save the current setting, affected pages and dates. Export the available visibility data and record qualified enquiries or orders over a period that fits your sales cycle.

2. Write down the reason.

   Is this a choice about how your content may be used, a response to a documented business loss, or a test of discovery value? Each calls for different evidence. You do not need to disguise an ownership principle as a traffic forecast.

3. Define what would change your mind.

   Name the business outcome, who will judge it and when. If you run a before-and-after comparison, record promotions, seasonality and other site changes. Treat the result as a signal with limitations, not automatic proof of cause.

4. Verify the actual change.

   Have the person responsible for the website confirm the selected control and its scope. Keep a record that allows you to restore the previous setting. Avoid bundling it with unrelated crawling or indexing changes.

A new report can help you ask a better question. Your customer records must help answer it.