Keep or Opt Out of Google Generative AI Search Features?

Most businesses should keep their site included in Google’s Search generative AI features by default. Opt out only when you have a documented reason—and only after you measure what inclusion is already doing for you in Search Console. This article walks through that decision with a scorecard, a 30-day diagnostic, and a clear view of what exclusion does and does not change.

Google’s June 2026 updates give website owners a Search Console control for AI Overviews, AI Mode, and generative AI features in Discover, plus dedicated impression reporting. The control exists. The harder question for owners and marketing leads is when flipping it is a rational business move versus a panic response to traffic anxiety.

Key takeaways

• Default to include. Google’s default Search generative AI control keeps your links and content eligible to appear in AI Overviews, AI Mode, and generative Discover features.

• Measure before you exclude. Use the Generative AI performance report (when available for your property) to see which pages earn AI-feature impressions, then compare that against business outcomes—not vanity traffic alone.

• Exclusion is not a ranking lever for classic results. Google states the control is not used as a ranking or inclusion signal for other parts of Search. Competitors can still appear in AI answers if you leave.

• Do not confuse this toggle with Google-Extended. They solve different problems: one governs Search generative AI appearance and grounding; the other helps limit certain model training and related uses.

What changed for website owners

According to Google’s Search Central documentation, AI Overviews and AI Mode surface supporting links drawn from the same core Search systems that power classic results. There are no extra technical requirements beyond being indexed and eligible for a normal Search snippet. Foundational SEO still matters: crawlability, helpful people-first content, textual clarity, and structured data that matches visible page text.

Separately, Google announced and documented a Search Console control that lets owners include or exclude their site’s links and content from Search generative AI features. Inclusion is the default. Exclusion means your content will not appear as links in those features and will not be eligible to help ground those AI responses. Google also notes that other sites’ content can still appear—and may look similar to yours.

Scale context matters for prioritization, not for panic. Google has publicly stated that AI Overviews reach over 2.5 billion monthly users and that AI Mode has surpassed 1 billion monthly users. That does not mean every query for every business triggers an AI Overview. Google’s own guidance says AI Overviews are shown when systems determine they are additive to classic Search—and often do not trigger.

The Inclusion Decision Scorecard

Use this six-factor scorecard before changing the control. Score each factor from 1 (weak / against inclusion) to 5 (strong / favors keeping inclusion). Total possible: 30.

How to read the total

• 23–30: Keep inclusion. Invest in stronger pages and measurement, not exclusion.

• 15–22: Keep inclusion for now. Run the 30-day diagnostic below, then reassess with evidence.

• 6–14: Exclusion may be justified for a defined period—but write down the hypothesis, the metric that would reverse it, and the review date.

A 30-day diagnostic before you touch the toggle

Google’s Help documentation recommends using the Generative AI performance report to understand how changing the control may affect traffic. The report (where rolled out) focuses on impressions in AI Overviews and AI Mode, with dimensions for pages, countries, devices, and dates. It is rolling out gradually; some properties will not see it yet.

1. Confirm your control status. In Search Console, open Settings → Search generative AI. Note whether the property is set to include, exclude, or inherit from a parent. Child properties can inherit parent settings, so check the property you actually manage.

2. Pull generative AI impressions by page (if the report is available). Export the top impression URLs. Mark each as commercial, brand, or informational support.

3. Compare against overall Web performance. Google notes that AI-feature appearances are included in overall Search Console Web search data. Look for pages where classic clicks fell while AI impressions rose—and pages where both declined (that pattern usually points to relevance or demand issues, not only AI Overviews).

4. Check quality of visits in Analytics. Google has stated that clicks from results pages with AI Overviews tend to be higher quality in the sense that users spend more time on site. If your engaged sessions and conversions from organic search remain healthy, a raw click decline may be less catastrophic than it looks.

5. Run a competitor presence check. For your top 10 commercial and comparison queries, note whether rivals appear as supporting links in AI Overviews or AI Mode when you do not. Exclusion does not empty the feature—it removes you from it.

6. Decide with a written hypothesis. Example: “We will exclude for 45 days because Page X’s informational traffic is unmonetizable and AI impressions concentrate there; we will reverse if branded demo requests drop more than Y%.” Without a reverse trigger, exclusion becomes permanent by inertia.

Worked example: regional home-services brand (hypothetical)

Consider a hypothetical multi-location HVAC company. Organic leads come from service pages (“furnace repair near me”) and a few educational posts (“heat pump vs furnace”). Over eight weeks, informational article sessions drop, while service-page leads are roughly flat. Leadership wants to “turn off AI Overviews.”

Scorecard application (hypothetical scores): measurement readiness 4, AI-feature evidence 3 (report available; educational URLs dominate impressions), business model fit 5 (leads matter more than article pageviews), competitive presence 5 (local competitors already appear in AI answers), content differentiation 3, legal constraint 5. Total: 25.

Decision: keep inclusion. The rational next step is not opt-out—it is upgrading commodity education into non-commodity pages (job photos, pricing ranges, failure diagnostics, technician checklists) and tightening conversion paths on pages that still earn clicks. Exclusion would mostly remove brand presence from AI answers while leaving rivals visible.

When exclusion can be a rational choice

Exclusion is typically more defensible when several of these are true at once:

• You have contractual or policy constraints against generative reuse of specific materials.

• Your primary commercial pages never appear in generative AI features, while unmonetizable pages dominate impressions and create brand or accuracy risk you cannot fix quickly.

• You have already tested content improvements and tracking, and leadership accepts that competitors will remain visible in AI features.

Google’s documentation is clear about timing: after you change the control, exclusion generally takes effect within a few days, with some content taking longer due to caching and propagation. Plan measurement windows accordingly. Also remember: this control does not replace a noindex directive, does not override Merchant Center or Google Ads participation, and does not by itself stop AI training—Google points publishers to Google-Extended for limiting certain model training uses.

Common failure patterns

• Panic opt-out after one bad week. Short windows mix seasonality, algorithm noise, and reporting lag. Prefer 28–45 days of paired Search Console and Analytics evidence.

• Treating exclusion as an SEO ranking tactic. Google states the control is not used as a ranking signal for other parts of Search. If classic rankings slipped, fix crawl, relevance, and page quality—not the generative AI toggle.

• Confusing Search generative AI control with Google-Extended. They address different surfaces and purposes. Misconfiguring either can create false confidence that you “solved AI.”

• Chasing AEO/GEO hacks while ignoring foundations. Google’s generative AI optimization guide says SEO fundamentals remain relevant, and it explicitly says you do not need special files such as llms.txt, “chunking” for AI, or special schema solely for AI Overviews on Google Search.

• Optimizing for impressions only. Impression growth without engaged visits or conversions is incomplete success. Pair generative AI impressions with lead and revenue outcomes.

What to do next

1. Complete the Inclusion Decision Scorecard with your team in one sitting.

2. Verify Search Console settings and, if available, export generative AI impressions by page.

3. Prioritize upgrading the top impression pages that map to revenue into non-commodity assets—original process, proof, pricing clarity, and specific expertise.

4. Only then consider exclusion, with a written hypothesis and a scheduled review.

If you need help connecting Search Console evidence to a practical SEO roadmap—and to broader AI-search visibility beyond Google—Oasbit’s search engine optimization services and generative engine optimization services are built for that measurement-first approach. For a broader channel sequence across website, ads, SEO, and GEO, see our guide on how to sequence those investments in the first 90 days.

If you want a working session on your Search Console data and whether inclusion or exclusion fits your model, book a growth strategy session.

Sources

• Google Search Central: AI features and your website

• Google Search Central: Optimizing for generative AI features on Google Search

• Search Console Help: Search generative AI control

• Search Console Help: Generative AI performance report (Search)

• Google Search Central Blog: Introducing Search Generative AI performance reports (June 3, 2026)

• Google Blog: New opportunities, control and insights for website owners (June 3, 2026)