Skip to content
← Guides & updates

SEO · FOR HONG KONG BUSINESSES

Google refreshed “Control what you share with Search”: which control should HK SMEs use?

On 21 October 2026 Google consolidated its Search sharing controls docs—noindex, robots meta / data-nosnippet, and the Search Console generative AI control. This guide helps Hong Kong SMEs match the tool to the goal and keep Google-Extended training controls separate.

Articles in this section are prepared with AI assistance. Sources are listed below.

On 21 October 2026, Google Search Central updated its “Control what you share with Google Search” documentation. The refresh gathers existing options in one place—including the robots meta tag, data-nosnippet, X-Robots-Tag, and Search Console’s Search generative AI control—and clarifies that the page applies to Google Search. Google’s documentation changelog describes this as keeping the page current with options site owners already have, not as a new ranking system.

Why this matters for Hong Kong SMEs

A typical small-business site mixes public service pages, staff login areas, quote PDFs, booking thank-you pages, and a separate question: whether content should appear in AI Overviews or AI Mode. Each goal needs a different control. Choosing the wrong one can hide an important page from Search, or create a false sense that “turning off AI” also blocks model training. This guide follows Google’s official wording so you can match the tool to the outcome.

Separate three different goals

  • Keep a page out of regular Google Search results (while still allowing people with the link to open it).
  • Keep the page searchable, but limit snippet text—or stop a passage or the whole page from being used as a direct input for AI Overviews and AI Mode.
  • Manage site-wide inclusion in Search generative AI features (AI Overviews, AI Mode, and generative AI features in Discover). This is separate from AI model-training controls.

Goal 1: Keep a page out of Search

Google’s guide says the strongest options are removing the content from your site or password-protecting private files. If the URL must stay open to people who have the link, but should not appear in Search, use noindex (robots meta tag or X-Robots-Tag). Googlebot must be able to crawl the URL to see that rule; blocking an HTML page in robots.txt can stop the crawler from discovering noindex. On the refreshed page, robots.txt is framed mainly as a way to stop crawling of image and video files.

  • Examples that often warrant password protection or noindex: staff dashboards, internal price lists, quote pages meant only for people with a private link.
  • Examples that should usually stay indexable: public service descriptions, opening hours, and how to enquire.
  • Content already showing in Search can also go through Google’s removal process; controls still take time to apply.

Goal 2: Control snippets and previews

To keep a page searchable while adjusting how Google presents it, use robots meta tags, data-nosnippet, or the X-Robots-Tag HTTP header. Google’s robots specifications state that nosnippet suppresses text snippets and video previews and also prevents that page’s content from being used as a direct input for AI Overviews and AI Mode; data-nosnippet can exempt only part of the HTML; max-snippet limits character length and likewise limits how much may be used as direct input for those AI features (with exceptions where a publisher has separately granted permission). These are page- or text-level presentation controls, not a whole-site exit from generative AI features.

Goal 3: Site-wide Search generative AI control

Since 31 August 2026, Settings > Search generative AI in Search Console is available to websites worldwide. Include is the default: your links and content may appear in AI Overviews, AI Mode, and generative AI features in Discover, may help ground responses, and may receive impressions and traffic from those features. Exclude prevents visibility and grounding in those features, and you will not receive traffic or impressions from them. Google states clearly that this control does not affect ordinary Google Search ranking or inclusion, and it is not an AI-training control.

After you change the setting, Google says exclusion generally takes a few days—often within 1–2 days after the control goes live, though some content can take longer because of caching. Child properties can inherit or override a parent. To monitor visibility, use the Generative AI performance report in Search Console separately.

Do not confuse this with Google-Extended

If your goal is to limit use of crawled content for training future Gemini models, or for grounding in Gemini Apps and related systems, look at the Google-Extended product token in robots.txt—not (or not only) the Search generative AI control. On 30 October 2026 Google also announced plans to support a shared AI-training robots.txt control (aimed at around 2027). Until that mechanism is widely available, Google-Extended remains the ready Google-specific training-related control. Search visibility, generative-AI feature appearance, and model training are three different levers.

Practical checklist for Hong Kong SMEs

  • List pages that must be findable in Search versus pages meant only for people with a link; consider passwords or noindex only for the second group.
  • Spot-check on a phone and in a private window: dashboards, thank-you pages, empty templates, and test folders should not appear in Search by mistake.
  • If a price, term, or personal detail should stay out of snippets, evaluate data-nosnippet or page-level nosnippet / max-snippet before excluding the whole site from generative AI features.
  • In Search Console, open Settings > Search generative AI, confirm Include or Exclude, and note the change date.
  • If training use matters to you, also review Google-Extended rules in robots.txt; do not assume the generative AI control covers training.
  • After changes, use the URL Inspection tool to confirm what Googlebot received, then re-check Search Console reports after a few days.

What this update is not

  • It is not a new core update or spam update.
  • Inclusion in AI features does not guarantee more enquiries; exclusion does not guarantee higher classic Search traffic.
  • The documentation applies to Google Search; other search engines or AI services may follow different rules.

Bottom line

The 21 October documentation refresh is useful as a decision map: removal or passwords, noindex, snippet controls, the Search generative AI control, and opt-outs for specific Google properties. Hong Kong SMEs can write down the outcome they want first, then pick the matching official tool. Clear, useful content and easy contact paths remain the shared foundation for Search and AI features; no setting is a ranking or business guarantee.

Sources

  1. Control what you share with Google Search (Google Search Central) ↗
  2. Latest Google Search documentation updates — October 2026 entry (21 Oct) ↗
  3. Robots meta tag, data-nosnippet, and X-Robots-Tag specifications ↗
  4. Block Search indexing with noindex (Google Search Central) ↗
  5. Search generative AI control — Search Console Help ↗
  6. AI features and your website (Google Search Central) ↗
  7. Google's common crawlers — Google-Extended product token ↗
  8. Google updates Control what you share doc (Search Engine Roundtable, 21 Oct 2026) ↗
WCAG 2.0Accessibility guidelines