Changelog
What's changing in search and AI discoverability - Google policy updates, new schema types, AI-crawler changes, and adoption opportunities. Pick an update on the right.
90 updates · last updated 2026-09-28
The Open Graph protocol specification documents a payment type namespace (https://ogp.me/ns/payment#) with properties including description, amount, expiration, status, identifier, and redirect URL. This is a new OG type for payment links. It is not yet a Google Search requirement.
What to do
No action needed for standard SEO. Monitor for adoption by social platforms and search engines before implementing.
The llms.txt v2 specification documents standard link relations: rel=describedby pointing to the llms.txt file, and rel=alternate type=text/markdown pointing to markdown versions of pages. These can be provided as HTML link elements or as HTTP Link response headers. These are ecosystem conventions for AI agent discoverability outside Google Search; Google Search does not use them specially.
What to do
Optionally add rel=describedby and rel=alternate type=text/markdown link relations to pages if you want to improve discoverability by non-Google AI agents and coding tools. Not required for Google Search.
DebugBear published a full implementation guide for WebMCP. WebMCP allows websites to expose structured tools that AI agents can discover and use directly instead of reverse-engineering the page interface via DOM inspection. Implementation requires server-side endpoint configuration and is not detectable by static file analysis.
What to do
Review the WebMCP implementation guide at https://www.debugbear.com/blog if your site has workflows that AI agents might need to perform (booking, purchasing, searching). No static-file change required.
Google Search Central changelog confirms additional supported properties were added to discussion forum structured data to provide more clarity on comment thread structure to Google ingestion systems. This prevents misinterpretations in handling of forum and Q&A content. No new schema type is required - only additional optional properties on existing DiscussionForumPosting markup.
What to do
Review the updated discussion forum structured data documentation and consider adding the newly supported properties to your DiscussionForumPosting markup to improve Google's understanding of your comment thread structure.
Google Search Central changelog confirms two new myths were added to the crawl budget documentation: (1) robots.txt is not a good way to control crawl budget, though it can be a method to indirectly free up crawl budget in the long run; (2) pages that serve non-200 status codes (except 304) do not waste crawl budget. These are clarifications to existing guidance, not new technical requirements.
What to do
No action needed. Review the updated crawl budget documentation to align your internal guidance with Google's clarified position on robots.txt and non-200 status codes as crawl budget factors.
Google introduced new performance metrics to surface how ads impact user experience. Real-user data for these metrics is available in the Chrome User Experience Report (CrUX) and Chrome DevTools. Site owners can now see ad impact metrics alongside Core Web Vitals data.
What to do
Review the new ad impact metrics in CrUX and Chrome DevTools to understand how ads on your site affect user experience. Optimize ad loading strategies if ad-related metrics are poor.
Apple Support documentation confirms that applying the nosnippet meta tag directive to Applebot prevents Apple AI models from using that content as additional context when generating output for display in Apple products and services (e.g., broad world knowledge answers in Siri and Search). This is distinct from the Google nosnippet behavior and is specific to Apple Intelligence features.
What to do
If you want to prevent your content from being used as context in Apple AI-generated answers (Siri, Search broad world knowledge), apply <meta name="robots" content="nosnippet" /> or <meta name="applebot" content="nosnippet" /> to the relevant pages. Note this will also prevent Apple from generating descriptions for those pages.
Apple Support documentation confirms that pages marked isAccessibleForFree:false in structured data are eligible to appear in Applebot search results but Applebot will not use that content as additional context when AI models generate output for Apple products. Section-level markup using hasPart is not supported by Applebot - only page-level isAccessibleForFree applies. This is a nuance distinct from Google's paywall handling.
What to do
If you have paywalled content and want to prevent it from being used as Apple AI context, add isAccessibleForFree:false at the page level in your JSON-LD structured data. Do not rely on hasPart section-level markup as Applebot does not support it.
W3C WCAG overview confirms ISO/IEC 40500:2026 incorporating the December 2024 update to WCAG 2.2 is expected to be published by late 2026. The 2026 version of EN 301 549 already uses WCAG 2.2. Organizations subject to the European Accessibility Act should ensure WCAG 2.2 compliance.
What to do
Organizations subject to the European Accessibility Act (EAA) or EN 301 549 should ensure their web content meets WCAG 2.2 success criteria. Content conforming to WCAG 2.2 also conforms to WCAG 2.1 and 2.0.
Cohere RAG documentation recommends breaking documents into chunks of approximately 400 words for best performance with Cohere models. This is a Cohere-specific API optimization recommendation and not a general SEO or web publishing requirement.
What to do
No action needed for standard web publishing. This recommendation applies only when using Cohere's Chat API with RAG and passing documents as context. Standard web pages do not need to be restructured to meet this guideline.
Google released a September 2026 Spam Update on or around September 24, 2026, as confirmed by Search Engine Roundtable. The update targets sites violating Google's spam policies. Sites experiencing ranking volatility around September 23 - 24, 2026 may be affected.
What to do
Audit your site against Google's spam policies at https://developers.google.com/search/docs/essentials/spam-policies. If you experienced ranking drops around September 23 - 24, 2026, review your content for scaled content abuse, cloaking, doorway pages, and other spam policy violations.
Moz Blog Whiteboard Friday (Crystal Carter) covers NLWeb (Natural Language Web), which uses the ASK protocol and structured data to prepare sites for the agentic web. NLWeb allows AI agents to interact with websites using natural language queries via a standardized protocol. This is an emerging standard distinct from WebMCP and UCP.
What to do
Monitor NLWeb and ASK protocol adoption. Review your structured data implementation to ensure it is comprehensive, as NLWeb leverages existing schema.org markup. No immediate implementation action is required as the protocol is emerging.
Bing Webmaster Blog confirms IndexNow now enables faster and more reliable content updates specifically for Shopping and Ads content on Bing. Sites using IndexNow can now benefit from near-real-time content discovery for product listings and ad-related pages, reducing the lag between content updates and Bing indexing.
What to do
If you operate an ecommerce or advertising site, implement IndexNow to notify Bing of product and content updates in real time. Submit IndexNow pings whenever product prices, availability, or ad landing pages change.
Bing Webmaster Blog published guidance on how AI-powered search is changing conversion measurement and attribution. Traditional last-click attribution models may undercount the influence of AI search interactions on conversions. Site owners should review their analytics setup to account for AI search touchpoints.
What to do
Review your conversion attribution model to ensure AI search interactions (Copilot citations, AI-generated answers) are captured. Consider multi-touch attribution models that account for AI search as an assist channel.
Google has updated VideoObject structured data documentation to add support for the creator property and to update guidance on the interactionStatistic property. The creator property allows video publishers to attribute video content to a specific person or organization, improving entity disambiguation and authorship signals for video content in Google Search.
What to do
Add the creator property to existing VideoObject structured data markup. Also review the updated interactionStatistic documentation to ensure your video interaction counts are correctly marked up with the supported interaction types.
W3C confirms that ISO/IEC 40500:2026, incorporating the December 2024 update to WCAG 2.2, is expected to be published by late 2026. The 2026 version of EN 301 549 (the European accessibility standard) already references WCAG 2.2. Organizations subject to the European Accessibility Act should ensure WCAG 2.2 conformance now.
What to do
Ensure your site conforms to WCAG 2.2 (the current W3C recommendation). If your organization is subject to the European Accessibility Act, use the WCAG 2.2 resources and EN 301 549 2026 as your compliance baseline. No action needed if already WCAG 2.2 compliant.
Apple Support confirms that the nosnippet meta tag directive applied to Applebot prevents Apple AI models from using that page's content as additional context when generating output for Apple products and services, including broad world knowledge answers in Siri and Search. This behavior is distinct from Google's nosnippet, which only prevents snippet display in search results.
What to do
If you want your content to be used as context by Apple AI models (Siri, Apple Intelligence), do not apply nosnippet to Applebot. If you wish to opt out of Apple AI context generation for specific pages, apply the nosnippet directive via meta tag or X-Robots-Tag HTTP header targeted at Applebot.
Apple Support confirms that pages marked isAccessibleForFree:false in structured data are eligible to appear in Applebot search results but Applebot will not use that content as additional context when AI models generate output for Apple products. Section-level hasPart markup is not supported by Applebot - only page-level isAccessibleForFree is recognized. This differs from Google's paywall handling which supports section-level markup.
What to do
If you have paywalled content and want it to appear in Apple Search results but not be used by Apple AI models, mark it with isAccessibleForFree:false at the page level in your JSON-LD. Do not rely on hasPart section-level markup for Applebot - it is not supported. To opt out of AI training use, disallow Applebot-Extended in robots.txt separately.
Cohere's RAG documentation recommends breaking documents into chunks of approximately 400 words for best retrieval performance with Cohere models. This is a Cohere-specific API optimization recommendation and not a general web publishing or SEO requirement. Google has explicitly stated that content chunking is not required for Google AI features.
What to do
No action needed for standard web publishing or Google SEO. If you are building a RAG pipeline using Cohere's API, follow their chunking guidance for optimal model performance.
Google Search Console Performance Report now includes a Multimodal Search Type filter, allowing site owners to segment performance data by multimodal search interactions. This is a reporting feature on the Google side and requires no publisher implementation.
What to do
No action needed. Review your Search Console Performance reports using the new Multimodal Search Type filter to understand how your content performs in multimodal search contexts.
Google has updated its hreflang documentation examples to use uppercase country codes (e.g., en-US instead of en-us) but has explicitly stated that it does not matter - both uppercase and lowercase country codes are accepted and treated identically. No code changes are required for existing implementations using lowercase codes.
What to do
No action needed. Existing hreflang implementations using lowercase country codes (e.g., en-us, fr-fr) continue to work correctly. You may optionally update to uppercase for consistency with Google's documentation examples, but this has no functional impact.
Bing Webmaster Blog confirms Bing now supports the data-nosnippet HTML attribute, which prevents specific sections of a page from being used in Bing search snippets. This is analogous to Google's existing data-nosnippet support and allows publishers to exclude specific content blocks from snippet generation on Bing.
What to do
If you have sections of content you do not want Bing to use in snippets, add the data-nosnippet attribute to the relevant HTML element: <div data-nosnippet>Content excluded from snippets</div>. This is now supported by both Google and Bing.
Bing Webmaster Blog published an article examining whether duplicate content hurts SEO and AI search visibility on Bing and Copilot. The article provides informational guidance on how Bing handles duplicate content in both traditional search and AI-generated answers. No new technical requirements are introduced.
What to do
Review the Bing Webmaster Blog article for guidance on how duplicate content may affect your visibility in Bing AI search features. Standard duplicate content best practices (canonical tags, consistent URL structures) remain applicable.
Bing Webmaster Blog published an article confirming that sitemaps remain an important mechanism for keeping content discoverable in AI-powered search on Bing. The article reinforces that submitting and maintaining accurate sitemaps helps Bing discover and index content for both traditional and AI search features.
What to do
Ensure your sitemap.xml is up to date, submitted in Bing Webmaster Tools, and referenced in your robots.txt. This is especially important for sites that publish or update content frequently and want to maximize discoverability in Bing AI search features.
Google Search Central changelog confirms VideoObject structured data documentation was updated to include support for the interactionStatistic property and to clarify supported interaction types. This allows publishers to document engagement signals such as view counts directly in their video structured data.
What to do
Review the updated VideoObject structured data documentation at https://developers.google.com/search/docs/appearance/structured-data/video and add interactionStatistic to your VideoObject markup where applicable. Example:
{
"@type": "InteractionCounter",
"interactionType": {
"@type": "WatchAction"
},
"userInteractionCount": 12345
}
Google Search Central changelog confirms additional supported properties were added to discussion forum structured data to provide more clarity on comment thread structure to Google ingestion systems, preventing misinterpretations in handling of forum and Q&A content. No new schema type is required; only additional optional properties on existing DiscussionForumPosting markup.
What to do
Review the updated discussion forum structured data documentation at https://developers.google.com/search/docs/appearance/structured-data/discussion-forum and add the newly supported properties to your DiscussionForumPosting markup where applicable to improve comment thread clarity for Google.
Google introduced new performance metrics in the Chrome User Experience Report (CrUX) and Chrome DevTools that surface how ads impact user experience. Real-user data for these ad impact metrics is now available alongside Core Web Vitals data. This extends the existing CrUX dataset with ad-specific performance signals.
What to do
Review the new ad impact metrics in CrUX and Chrome DevTools to understand how advertising on your pages affects user experience. Optimize ad loading strategies if ad-related metrics show poor scores.
DebugBear published a full implementation guide for WebMCP. Instead of forcing AI agents to reverse-engineer a website's interface via DOM inspection or screenshots, WebMCP allows websites to expose structured tools that agents can discover and use directly. This is an emerging protocol for AI agent discoverability that complements existing DOM/accessibility-tree-based agent access.
What to do
Review the WebMCP implementation guide at https://www.debugbear.com/blog if your site handles tasks that AI agents might perform on behalf of users (booking, purchasing, searching). WebMCP requires server-side endpoint configuration and is not a static-file change.
The llms.txt v2 specification documents significant adoption: thousands of sites publish llms.txt files, Chrome Lighthouse 13.3 introduced an Agentic Browsing audit category that checks for llms.txt, AI labs publish their own llms.txt files, and major documentation platforms generate one automatically. Google Search itself does not use llms.txt in a special way, but the broader AI agent ecosystem (coding agents, chat assistants, documentation tools) actively uses it for content discovery.
What to do
Consider publishing an llms.txt file at your site root if you want your content to be more accessible to AI coding agents, chat assistants, and documentation tools outside of Google Search. This has no Google Search ranking impact but may improve discoverability in non-Google AI contexts. Follow the v2 spec format: H1 site name, blockquote summary, optional detail sections, H2-delimited file lists with markdown links.
Google has added a new guide explaining how site owners can add a Search profile badge to their website to help point their audience to their Google Search profile. This is a new Google Search feature for publishers and brands.
What to do
Review the new guide at https://developers.google.com/search/docs/appearance/search-profile-badge to understand eligibility and implementation steps for adding a Search profile badge to your site.
Google has updated the aggregator unit and supplier unit search features to support local business queries. This extends the regional search experience documentation with confirmation that local business queries are now eligible for these units in supported countries.
What to do
Review the regional differences in Search experience documentation at https://developers.google.com/search/docs/appearance/regional-search-experience to understand eligibility criteria for aggregator and supplier units in your target markets.
Google Search Central changelog confirms additional supported properties were added to discussion forum structured data to provide more clarity on comment thread structure to Google ingestion systems. This prevents misinterpretations in Google's handling of forum and Q&A content. The update adds optional properties to existing DiscussionForumPosting markup rather than introducing a new schema type.
What to do
Review the updated discussion forum structured data documentation and consider adding the newly supported properties to your DiscussionForumPosting markup if your site hosts forum or Q&A content. This can improve how Google interprets threaded comment structures on your pages.
Google added two new myth-busting notes to the crawl budget documentation: (1) robots.txt is not a good way to directly control crawl budget, though it can indirectly free up crawl budget in the long run; (2) pages that serve non-200 HTTP status codes (except 304) do not waste crawl budget. These are documentation clarifications with no change in Google's actual crawling behavior.
What to do
No action needed. If you have been using robots.txt as a primary crawl budget management tool, review Google's crawl budget optimization guide for more effective approaches. Do not be concerned that error pages are consuming crawl budget.
Google Search Central published a new guide explaining how site owners can add a Search profile badge to their website to help point their audience to their Google Search profile. This is a Google-side feature that requires having a Google Search profile configured. No structured data or meta tag change is required.
What to do
If you have a Google Search profile and want to help your audience find it, review the new guide at https://developers.google.com/search/docs/appearance/search-profile-badge for implementation instructions.
Google Search Central added two new myths to the crawl budget documentation. First, robots.txt is not a good way to control crawl budget, though it can indirectly free up crawl budget in the long run. Second, pages that serve non-200 HTTP status codes (except 304) do not waste crawl budget. These clarifications correct common misconceptions about crawl budget management.
What to do
Do not rely on robots.txt as a primary crawl budget management tool. Do not worry that error pages (4xx, 5xx) are consuming crawl budget - they do not. Focus on crawl budget optimization through internal linking, sitemap quality, and reducing duplicate content instead.
The Chrome User Experience Report (CrUX) now includes ad-related metrics in addition to Core Web Vitals data. DebugBear confirms Google introduced new performance metrics to surface how ads impact user experience, available in CrUX and Chrome DevTools. This is a data reporting expansion, not a new ranking signal.
What to do
Review your CrUX data in PageSpeed Insights and Search Console with awareness that ad-related performance metrics are now included. Use Chrome DevTools to inspect how ads on your pages affect user experience metrics.
The W3C WCAG overview page confirms that ISO/IEC 40500:2025 is exactly the same as the October 2023 version of WCAG 2.2, and that the December 2024 version of WCAG 2.2 is expected to be published as ISO/IEC 40500:2026 by late 2026. The 2026 version of the European Standard EN 301 549 (used to address the European Accessibility Act) now references WCAG 2.2. WCAG 2.2 adds 9 new success criteria over WCAG 2.1, and success criterion 4.1.1 Parsing is obsolete in WCAG 2.2.
What to do
Ensure your site conforms to WCAG 2.2 (the latest version) rather than WCAG 2.1 or 2.0. If your site must comply with the European Accessibility Act (EAA) via EN 301 549, the 2026 version of EN 301 549 references WCAG 2.2 - audit against WCAG 2.2 success criteria. You do not need to separately audit against WCAG 2.1 or 2.0 as WCAG 2.2 is backwards compatible.
Google's 'Creating helpful, reliable, people-first content' guide explicitly states that E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) itself is not a specific ranking factor. Google's automated systems use a mix of factors that can identify content demonstrating good E-E-A-T, but there is no single E-E-A-T score or signal. Trust is identified as the most important of the four aspects. YMYL (Your Money or Your Life) topics receive additional weight for E-E-A-T signals.
What to do
Focus on creating genuinely helpful, people-first content with clear authorship (bylines, author pages), transparent content creation processes (including AI disclosure where relevant), and first-hand expertise. Do not attempt to game a specific E-E-A-T metric - there is none. Use the Who/How/Why framework from Google's guide to self-assess content quality.
Google's E-E-A-T guide states that AI or automation disclosures are useful for content where someone might think 'How was this created?' and recommends adding these disclosures when they would be reasonably expected. The guide asks whether the use of automation or AI-generation is self-evident to visitors through disclosures, whether background is provided about how automation was used, and whether the reason automation was seen as useful is explained. Using automation to produce content primarily to manipulate search rankings is a violation of Google's spam policies.
What to do
Add clear disclosures when AI or automation is used to substantially generate content, particularly for content types where readers would expect to know the creation process (e.g., product reviews, news articles, research). Ensure disclosures explain what role AI played and why it was used. Do not use AI generation primarily to manipulate search rankings.
Apple's Applebot documentation confirms that applying the nosnippet robots directive to a page prevents Applebot from using that page's content as additional context when Apple AI models generate output for display in Apple products and services (e.g., Siri, Apple Intelligence). This is distinct from the Applebot-Extended opt-out for training data. The nosnippet directive affects real-time AI-generated answers, while Applebot-Extended controls training data inclusion.
What to do
If you want to prevent your content from being used as context in Apple AI-generated answers (e.g., Siri broad world knowledge answers) without blocking indexing entirely, apply the nosnippet directive: <meta name="robots" content="nosnippet" /> or <meta name="applebot" content="nosnippet" />. If you want to prevent training data use only, disallow Applebot-Extended in robots.txt instead.
Apple's Applebot documentation confirms that if a site's robots.txt does not mention Applebot but does mention Googlebot, Applebot will follow the Googlebot instructions. This means sites that have only configured robots.txt for Googlebot are implicitly also configuring Applebot's crawl behavior. Applebot does not follow crawl-delay directives.
What to do
If you want Applebot to behave differently from Googlebot, add explicit Applebot directives to your robots.txt. If you are comfortable with Applebot following your Googlebot rules, no action is needed. Be aware that blocking Googlebot in robots.txt will also block Applebot unless you add an explicit Applebot allow rule.
Google now requires that adult products be explicitly labeled in ecommerce feeds. This is a mandatory requirement distinct from the existing has-adult-consideration-schema rule (which covers the schema.org hasAdultConsideration property on structured data). Failure to label adult products in feeds may result in policy violations, removal of product listings, or account-level penalties in Google Merchant Center. This requirement applies to all merchants submitting product feeds to Google that include adult-oriented products. The labeling must be applied at the feed level in addition to any structured data markup on the page.
What to do
In your Google Merchant Center product feed, add the 'adult' attribute with value 'yes' for all adult-oriented products: <g:adult>yes</g:adult> (for XML feeds) or set the 'adult' column to 'yes' (for spreadsheet feeds). Also ensure your product pages include the hasAdultConsideration property in structured data:
{
"@type": "Product",
"hasAdultConsideration": true
}
. Review Google Merchant Center's adult content policy at https://support.google.com/merchants/answer/6150138 for the full list of product categories that require this label.
Google has dropped free product listings from Google Search results in the European Economic Area (EEA). Ecommerce sites that previously relied on free product listings (unpaid Shopping results) for visibility in EEA markets will no longer receive this traffic. This change is likely related to Google's compliance with EU Digital Markets Act (DMA) obligations and the ongoing Ad Tech antitrust regulatory environment in Europe. Merchants targeting EEA markets must now evaluate paid Shopping campaigns or other organic search strategies to maintain product visibility in those markets. This is distinct from the existing site-reputation-policy-eea rule which covers content policy enforcement adjustments.
What to do
Review your Google Search Console Performance reports and Google Merchant Center data to assess the impact of free product listing removal in EEA markets (EU member states, Iceland, Liechtenstein, Norway). Consider: (1) evaluating Google Shopping paid campaigns for EEA markets to replace lost free listing visibility; (2) strengthening organic product page SEO (Product structured data, page experience, content quality) to maintain organic web result visibility; (3) reviewing your Merchant Center account settings for EEA-specific configuration options; (4) monitoring Google Search Central and Merchant Center announcements for any alternative free listing programs or DMA compliance updates that may affect product visibility in European markets.
Google now supports local business queries in aggregator and supplier search units, extending these rich result experiences to local business verticals. This builds on the existing regional-google-search-experience-documentation rule which documents that aggregator units, supplier units, and carousels are available in certain countries. Local businesses in supported regions and verticals should ensure their structured data and Google Business Profile are correctly configured to be eligible for these units. Aggregator units typically appear for queries like 'restaurants near me' or 'plumbers in [city]' and pull from aggregator platforms; supplier units show individual business listings. Correct LocalBusiness structured data and a verified Google Business Profile are foundational requirements for eligibility.
What to do
To maximize eligibility for local aggregator and supplier search units: (1) ensure your Google Business Profile is verified and fully completed with accurate name, address, phone, hours, and category; (2) add LocalBusiness structured data to your website:
{
"@context": "https://schema.org",
"@type": "LocalBusiness",
"name": "Business Name",
"address": {
"@type": "PostalAddress",
"streetAddress": "123 Main St",
"addressLocality": "City",
"addressRegion": "State",
"postalCode": "12345",
"addressCountry": "US"
},
"telephone": "+1-555-555-5555",
"openingHours": "Mo-Fr 09:00-17:00"
}
; (3) ensure your NAP (Name, Address, Phone) is consistent across your website, Google Business Profile, and any aggregator platforms; (4) review Google's regional search experience documentation for the specific aggregator and supplier unit verticals and regions supported in your target markets.
Google has added a new guide on how to add a Search profile badge to a website. The Search profile badge allows site owners to point their audience to their Google Search profile, helping users find and follow the site's presence on Google Search. This is a new feature announced in the Google Search Central documentation changelog (last updated 2026-09-16 UTC). Site owners who want to help their audience discover their Search profile should implement the badge on their site.
What to do
Review Google's guide on adding a Search profile badge at https://developers.google.com/search/docs/appearance/search-profile-badge. Implement the badge on your site following the instructions provided to help your audience find and follow your Search profile.
Site rules set for Applebot-Extended in robots.txt are not considered in ranking for Google Search. This is explicitly confirmed in Apple's own documentation: 'Site rules for Applebot-Extended are not considered in ranking for Search.' Applebot-Extended is Apple's secondary user agent used to determine how data crawled by the main Applebot user agent can be used - specifically for training Apple's general purpose foundation models powering Apple Intelligence, Services, and Developer Tools. Critically, Applebot-Extended does NOT crawl webpages itself; it only determines how to use data already crawled by Applebot. Disallowing Applebot-Extended prevents your content from being used to train Apple foundation models, but pages that disallow Applebot-Extended can still be included in Spotlight, Siri, and Safari search results. Directives targeting Applebot-Extended (e.g., disallow rules in robots.txt) have no effect on how Google crawls, indexes, or ranks your pages. Site owners should not assume that blocking or allowing Applebot-Extended will influence their Google Search performance. Manage Applebot-Extended rules separately from your Google Search crawling strategy.
What to do
Do not rely on Applebot-Extended robots.txt rules to influence Google Search ranking or crawling. If you have robots.txt rules targeting Applebot-Extended, review them independently of your Google Search crawling strategy. To opt out of Apple foundation model training while retaining Spotlight/Siri/Safari search visibility: User-agent: Applebot-Extended Disallow: / This will not affect your Google Search rankings or Applebot's ability to index your pages for Apple search products. Ensure your Googlebot directives are correctly configured separately from any Applebot-Extended directives. See the applebot-crawl-eligibility rule for full Applebot implementation guidance.
Google has launched a European Ecosystem Carousel feature in European search results, covering query types including weather, translations, and additional verticals. This is a Google-side feature for European markets and is distinct from the structured data carousels (beta) available in South Africa and Turkey. The European Ecosystem Carousel does not appear to require publisher-side structured data implementation - it is driven by Google's own data sources and ecosystem partners. Site owners targeting European markets should be aware of this feature as it may affect SERP layout and click-through rates for queries in covered verticals.
What to do
No structured data implementation required for the European Ecosystem Carousel. Monitor your Google Search Console Performance reports for European markets to understand how this carousel feature affects impressions and clicks for weather, translation, and other covered query types. If you operate a site in a covered vertical (weather, translations, etc.) in European markets, review Google's regional search experience documentation for any publisher eligibility requirements that may be added in future.
Google's Chrome User Experience Report (CrUX) now includes ad-related metrics alongside Core Web Vitals and other page experience data. Site owners and SEO professionals who use CrUX data for performance benchmarking and Core Web Vitals analysis should be aware that the dataset now encompasses ad performance signals. This may affect how CrUX data is interpreted in tools that surface it, including Google Search Console, PageSpeed Insights, and third-party SEO platforms. Sites that serve ads should monitor whether ad-related metrics in CrUX affect their page experience signals or how their performance data is reported.
What to do
No immediate code change required. Review your CrUX data in Google Search Console and PageSpeed Insights with awareness that ad metrics are now included in the dataset. If your site serves ads, audit your ad implementation for performance impact: ensure ads do not cause Cumulative Layout Shift (CLS), do not block Largest Contentful Paint (LCP), and do not degrade Interaction to Next Paint (INP). Use the web-vitals JavaScript library to measure the impact of ad loading on your Core Web Vitals scores.
Schema.org version 30.1 added two significant areas of new vocabulary: (1) Common eCommerce product and retail feed data - new properties for ShippingRateSettings and PriceTypeEnumeration, and added OfferShippingDetails as an existing property for retail feed alignment; (2) EU Digital Product Passport (DPP) discovery vocabulary - new class DigitalProductPassport with properties hasDigitalProductPassport, EnvironmentalProductDeclaration, DeclarationOfConformity, authorizedRepresentative, and recycledContentPercentage. The EU DPP vocabulary supports discovery of Digital Product Passports as required under the EU Ecodesign for Sustainable Products Regulation (ESPR). Additionally, version 30.1 corrected and expanded Italian non-profit enumerations and added explicit support for ordered lists. Ecommerce sites targeting EU markets should evaluate the new DPP vocabulary for compliance and structured data alignment. Sites using retail feed data should review the new retail feed properties for alignment with Merchant Center feed specifications.
What to do
Review schema.org version 30.1 release notes at https://schema.org/docs/releases.html. For ecommerce sites: (1) evaluate the new retail feed data properties (ShippingRateSettings, PriceTypeEnumeration, OfferShippingDetails) for alignment with your Merchant Center feed and product structured data; (2) for EU market sites, evaluate the new DigitalProductPassport class and associated properties (hasDigitalProductPassport, EnvironmentalProductDeclaration, DeclarationOfConformity, authorizedRepresentative, recycledContentPercentage) for EU ESPR compliance and structured data implementation. Example DPP markup:
{
"@type": "Product",
"name": "Product Name",
"hasDigitalProductPassport": {
"@type": "DigitalProductPassport",
"recycledContentPercentage": 30
}
}
. Monitor Google Search Central for any announcements about Google Search support for these new vocabulary types.
Botify has announced powerful upgrades to its MCP (Model Context Protocol) integration, enabling SEO data and AI agents to operate directly inside ChatGPT, Claude, and other major AI platforms. This confirms that MCP-based integrations are now production-ready in major AI tools, not just an emerging standard. SEO professionals and site owners should be aware that AI agents can now access, analyze, and act on SEO data (crawl data, log files, performance metrics) directly within AI chat interfaces via MCP. This extends the existing webmcp-protocol rule with confirmation that MCP integrations are live in production AI platforms. Sites that want to maximize AI agent discoverability and enable AI-assisted SEO workflows should evaluate MCP implementation as a near-term priority rather than a future consideration.
What to do
Evaluate MCP (Model Context Protocol) implementation for your site and SEO tooling: (1) review the Botify MCP documentation at https://www.botify.com/blog for implementation guidance on connecting your SEO data to AI platforms via MCP; (2) assess whether implementing a WebMCP endpoint (see webmcp-protocol rule) would enable AI agents to discover and interact with your site's structured data and services; (3) for SEO teams, explore MCP-enabled SEO tools that allow AI agents in ChatGPT and Claude to access your site's crawl data, performance metrics, and structured data directly; (4) monitor the Model Context Protocol specification at https://modelcontextprotocol.io for web-specific implementation guidance.
Apple's Applebot crawler powers Spotlight, Siri, Safari search, and Apple Intelligence features. Key implementation requirements: (1) Applebot respects standard robots.txt directives targeted at 'Applebot'; if robots.txt does not mention Applebot but mentions Googlebot, Applebot will follow Googlebot instructions. (2) Applebot supports robots meta tags in the <head> section: noindex (prevents indexing in Spotlight/Siri), nosnippet (prevents description/web answer generation AND prevents content from being used as additional context in Apple AI-generated output), nofollow, none, and index. (3) Applebot supports X-Robots-Tag HTTP response headers for non-HTML resources. (4) Applebot may render pages in a browser - JavaScript, CSS, and XHR must not be blocked via robots.txt for proper rendering. (5) For paywalled content, use isAccessibleForFree:false in JSON-LD structured data - such pages remain eligible for search results but Applebot will not use that content as additional context for AI-generated output. Section-level hasPart markup is not supported by Applebot (page-level only). (6) Applebot-Extended is a secondary user agent that does not crawl webpages - it only determines how to use data already crawled by Applebot; disallowing Applebot-Extended opts out of Apple foundation model training but does not affect Spotlight, Siri, or Safari search inclusion. (7) Applebot does not follow crawl-delay. (8) The iTMS user agent (for Apple Podcasts) does not follow robots.txt. This rule extends the existing applebot-extended-not-a-ranking-signal rule with full Applebot implementation guidance.
What to do
To configure Applebot crawling: (1) Add Applebot-specific robots.txt rules if needed:
User-agent: Applebot
Disallow: /private/
(2) To opt out of Apple AI training data collection:
User-agent: Applebot-Extended
Disallow: /
(3) Add robots meta tags in <head> for page-level control:
<meta name="robots" content="noindex" /> - prevents Spotlight/Siri indexing
<meta name="robots" content="nosnippet" /> - prevents AI context use
<meta name="applebot" content="nosnippet" /> - Applebot-specific nosnippet
(4) For paywalled content, add to JSON-LD:
{
"@context": "https://schema.org",
"isAccessibleForFree": false
}
(5) Ensure JS, CSS, and XHR are not blocked by robots.txt so Applebot can render pages properly. Verify Applebot traffic via reverse DNS lookup on *.applebot.apple.com or by checking against Apple's published CIDR prefix JSON at https://support.apple.com/en-us/HT204683.
The llms.txt specification has reached v2 with significant ecosystem adoption: thousands of sites publish an llms.txt file, major documentation platforms (Mintlify, GitBook, Wix, and others) generate one automatically, Chrome Lighthouse 13.3 now audits sites for llms.txt as part of its new 'Agentic Browsing' category, and AI labs (OpenAI, Anthropic, etc.) publish their own llms.txt files for their developer documentation. The llms.txt format provides a curated, LLM-readable overview of a site's content with links to detailed Markdown versions of pages, intended for use by AI coding agents, chat assistants, and other LLM-powered tools when they need to understand a site's content. IMPORTANT: This rule does NOT contradict the existing llms-txt-not-required rule. Google Search does not use llms.txt in a special way for Search ranking or AI Overviews/AI Mode inclusion. However, llms.txt is now a de facto standard for AI agent discoverability in non-Google contexts (coding agents, ChatGPT browsing, Perplexity, Claude, etc.), and Chrome Lighthouse now audits for it. Site owners - especially those with developer documentation, APIs, or technical content - should evaluate whether publishing an llms.txt file is appropriate for their AI agent discoverability strategy outside of Google Search.
What to do
To implement llms.txt: (1) create a file named llms.txt at your site root (https://yourdomain.com/llms.txt) or at any subpath to cover pages under that path; (2) follow the v2 spec format: start with an H1 (site/project name), add an optional blockquote summary, add optional detail sections, then add H2-delimited file lists with links to Markdown versions of key pages; (3) optionally provide Markdown versions of pages at the same URL with .md extension or ?format=markdown parameter; (4) add link relations in HTML head: <link rel="describedby" href="/llms.txt" /> and <link rel="alternate" type="text/markdown" href="/page.html.md" />; (5) or add as HTTP Link response header: Link: </docs/llms.txt>; rel="describedby". Many CMS and documentation platforms (Mintlify, GitBook, Wix, WordPress plugins, VitePress, Docusaurus) can generate llms.txt automatically. Reference: https://llmstxt.org/
Chrome Lighthouse 13.3 introduced a new 'Agentic Browsing' audit category that evaluates how well a website is prepared for AI agent interactions. DebugBear has confirmed this new category exists and includes audits such as checking for an llms.txt file. This is a new Lighthouse category separate from Performance, Accessibility, Best Practices, and SEO. As AI agents increasingly interact with websites on behalf of users (booking appointments, comparing products, searching documentation), Lighthouse is beginning to surface agentic readiness as a measurable quality signal. While Agentic Browsing scores do not currently feed into Google Search ranking signals, Lighthouse audits in this category represent emerging best practices that site owners - especially those with developer documentation, ecommerce, or booking functionality - should be aware of and begin addressing.
What to do
Run Lighthouse 13.3 or later on your site and review the new 'Agentic Browsing' category results. Key audits to address include: (1) publishing an llms.txt file at your site root (see llms-txt-v2-spec rule); (2) ensuring your DOM structure is clean and semantically meaningful for agent parsing; (3) ensuring your accessibility tree is well-formed with proper ARIA labels and roles; (4) ensuring critical actions are accessible without JavaScript-only interactions where possible. Monitor Google Chrome release notes and DebugBear blog (https://www.debugbear.com/blog) for updates to the Agentic Browsing audit category as it evolves.
The Open Graph protocol specification explicitly states: 'If the page specifies an og:image it should specify og:image:alt.' The og:image:alt property provides a text description of the og:image, serving the same purpose as the HTML alt attribute for social sharing contexts. This improves accessibility for users of assistive technologies when content is shared on social platforms, and provides context to social platforms about the image content. This is a distinct requirement from the existing missing-og-image rule (which checks for the presence of og:image) and the image-missing-alt rule (which checks for alt on HTML img elements).
What to do
Add og:image:alt alongside your og:image tag: <meta property="og:image:alt" content="A description of what is in the image" /> inside <head>. The alt text should describe the image content, not be a caption. Example: <meta property="og:image" content="https://yourdomain.com/og-image.jpg" /><meta property="og:image:alt" content="Screenshot of the product dashboard showing analytics charts" />
Chrome 151 (released August 2026) introduced support for measuring Core Web Vitals - including Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS) - for soft navigations in single-page applications (SPAs). Previously, Core Web Vitals metrics for SPAs only described the initial page load; subsequent in-app navigations were either attributed to the original URL or not measured at all. With Chrome 151, soft navigation reporting allows accurate per-route Core Web Vitals measurement for SPAs. DebugBear confirms: 'Soft navigation reporting lets you measure LCP and other metrics for the dynamic URLs used in single page applications (SPAs). Support for this shipped in Chrome 151 in August 2026.' SPAs that do not implement soft navigation reporting will have incomplete Core Web Vitals data in field tools (CrUX, Search Console), potentially masking performance regressions on specific routes. As Google's ranking systems use field Core Web Vitals data, incomplete measurement can lead to suboptimal optimization decisions.
What to do
Implement soft navigation reporting for your SPA: (1) update the web-vitals JavaScript library to the latest version which supports soft navigation metrics; (2) use the onLCP, onINP, and onCLS functions with the reportAllChanges option to capture metrics for soft navigations; (3) ensure your SPA uses the History API (pushState/replaceState) for navigation so Chrome can detect soft navigations; (4) verify soft navigation data is appearing in your analytics by checking for the navigationType field in web-vitals reports; (5) monitor Search Console Core Web Vitals report for per-URL data once soft navigation data propagates to CrUX. Example: import { onLCP, onINP, onCLS } from 'web-vitals'; onLCP(sendToAnalytics,
{ reportAllChanges: true }
); Reference: https://web.dev/soft-navigations/
Google has published a dedicated documentation page on regional differences in the Google Search experience, covering search experiences available in certain countries such as aggregator units, supplier units, and carousels. Search features, rich result types, AI features (AI Overviews, AI Mode), carousel structured data, and other Search experiences may vary by country or locale. The documentation is intended to help publishers, businesses, and aggregators learn about the different regional search features available and understand the eligibility criteria and how to implement them. Site owners targeting specific regions should review Google's region-specific Search experience documentation to understand which features are available in their target markets and ensure their structured data and technical SEO implementation is aligned with regional availability. For example, structured data carousels are available in South Africa and Turkey but not all regions; preferred sources availability varies by language; and AI Mode rollout is region-dependent.
What to do
Review Google's dedicated regional differences in Search experience documentation at https://developers.google.com/search/docs/appearance/regional-search-experience for regional availability of features relevant to your target markets. Check: (1) which rich result types are supported in your target regions; (2) whether AI Overviews and AI Mode are available in your target regions; (3) whether structured data carousel features (aggregator units, supplier units) are available in your target regions; (4) whether preferred sources is available in your target languages. Adjust your structured data implementation priorities based on regional feature availability. Monitor Google Search Central documentation and changelog for regional expansion announcements.
OpenAI uses OAI-AdsBot to validate the safety of web pages submitted as ads on ChatGPT. When a page is submitted as a ChatGPT ad, OpenAI may visit the landing page to ensure it complies with OpenAI's advertising policies. OpenAI may also use content from the landing page to determine when it is most relevant to show the ad to users. Key characteristics: (1) OAI-AdsBot only visits pages that have been submitted as ads - it does not crawl the web generally; (2) data collected by OAI-AdsBot is NOT used to train generative AI foundation models; (3) OAI-AdsBot is entirely independent of OAI-SearchBot (which controls ChatGPT Search eligibility) and GPTBot (which controls training data collection). Site owners running ChatGPT ad campaigns should ensure their ad landing pages are accessible to OAI-AdsBot and comply with OpenAI's advertising policies. Blocking OAI-AdsBot may affect ad policy validation and ad relevance scoring.
What to do
Ensure your ChatGPT ad landing pages are not blocked by robots.txt for OAI-AdsBot. To explicitly allow it: User-agent: OAI-AdsBot Allow: / Verify your landing pages comply with OpenAI's advertising policies. Published IP addresses for OAI-AdsBot are available at https://openai.com/adsbot.json. The full user-agent string is: Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; OAI-AdsBot/1.0; +https://openai.com/adsbot
Botify analysis of over 7 billion log file entries confirms that OpenAI has tripled its crawl of the web, and AI bots are significantly increasing their crawl activity across ecommerce and content sites. Botify has published a 'Site Readiness Checklist' for whether AI bots can pull product information from ecommerce sites. Key findings indicate that many ecommerce sites are inadvertently blocking AI crawlers from accessing product data through robots.txt rules, JavaScript-only rendering of product information, or incomplete structured data. As AI shopping agents increasingly source product information directly from crawled pages rather than relying on user-initiated searches, ensuring AI crawlers can access complete, accurate product data is becoming a critical visibility requirement. This extends the existing acp-ucp-agentic-commerce-protocols and agent-friendly-website rules with specific product data accessibility requirements.
What to do
Audit your site's AI crawler accessibility using the following checklist: (1) review robots.txt to ensure AI crawlers (GPTBot, OAI-SearchBot, PerplexityBot, ClaudeBot, and other AI crawlers) are not inadvertently blocked from product pages; (2) ensure product information (name, price, availability, description, images) is present in the initial server-rendered HTML, not injected via JavaScript after page load; (3) verify complete Product structured data (JSON-LD) is present on all product pages including price, availability, SKU, and images; (4) ensure product images are crawlable and have descriptive alt text; (5) check that pagination and category navigation is crawlable so AI bots can discover your full product catalog; (6) monitor your server logs for AI bot crawl activity and ensure your server can handle increased AI crawler traffic without rate-limiting or blocking. Reference Botify's feed readiness checklist at https://www.botify.com/blog for a complete implementation guide.
Google has confirmed that it can lose faith in an entire site based on low-value programmatic SEO. When a site produces large volumes of programmatically generated pages that provide little or no unique value to users, Google may reduce trust in the site as a whole - not just the individual low-quality pages. This can result in sitewide ranking suppression beyond what would be expected from individual page-level quality signals. This extends Google's scaled content abuse spam policy: even if programmatic pages are not explicitly spam, producing low-value templated content at scale can erode Google's confidence in the entire domain. Sites should ensure that programmatically generated pages provide genuine, unique value and are not simply thin variations of the same content.
What to do
Audit all programmatically generated pages on your site. For each template or content generation pattern, ask: does each generated page provide unique, substantive value that a user would find helpful? If not, consider: (1) consolidating thin pages into fewer, richer pages; (2) adding noindex to low-value generated pages; (3) blocking low-value generated URL patterns in robots.txt; (4) improving the content generation process to add genuine unique value per page. Monitor Google Search Console for sitewide traffic drops that may indicate trust loss from low-value programmatic content.
WebMCP (Web Model Context Protocol) is an emerging protocol that extends the Model Context Protocol (MCP) to the web, enabling AI agents to discover and interact with web services in a structured, machine-readable way. Botify has identified WebMCP as a relevant emerging standard for sites seeking to maximize discoverability by AI agents and LLM-powered tools. WebMCP allows AI agents to understand what actions and data a website exposes without relying solely on HTML parsing or visual rendering. This complements the existing acp-ucp-agentic-commerce-protocols rule and the agent-friendly-website rule. While WebMCP is not yet a formal ranking signal, early awareness and preparation positions sites for AI agent-driven discovery as the protocol matures.
What to do
Monitor the WebMCP specification and adoption timeline. Review the Botify blog post 'What Is WebMCP and Why Does It Matter?' for implementation guidance as it becomes available. For sites that expose APIs or structured data services, evaluate whether implementing a WebMCP endpoint would improve discoverability by AI agents. Ensure your existing structured data (JSON-LD, Open Graph) is complete and accurate as a foundation for any future WebMCP implementation.
OpenAI uses GPTBot to crawl web content that may be used in training its generative AI foundation models. Disallowing GPTBot in robots.txt signals that a site's content should not be used for AI model training. This directive is entirely independent of OAI-SearchBot (which controls ChatGPT Search eligibility) - a site can allow OAI-SearchBot to appear in ChatGPT search results while simultaneously disallowing GPTBot to opt out of training data collection. If your site has allowed both OAI-SearchBot and GPTBot, OpenAI may use results from a single crawl for both purposes to avoid duplicative crawling. Site owners should make a deliberate, explicit choice about whether to allow or disallow GPTBot based on their content licensing and AI training consent policies.
What to do
Make an explicit robots.txt decision about GPTBot based on your content policy: To opt OUT of AI training data collection: User-agent: GPTBot Disallow: / To opt IN (allow training): User-agent: GPTBot Allow: / To allow search visibility but opt out of training: User-agent: OAI-SearchBot Allow: / User-agent: GPTBot Disallow: / Review OpenAI's published IP ranges at https://openai.com/gptbot.json.
OpenAI uses the ChatGPT-User agent for user-initiated actions in ChatGPT and CustomGPTs - for example, when a user asks ChatGPT a question that causes it to visit a web page, or when users interact with external applications via ChatGPT plugins/tools. Because these actions are initiated by a user rather than automated crawling, robots.txt rules may not apply to ChatGPT-User. Critically, ChatGPT-User is NOT used to determine whether content appears in ChatGPT Search - search eligibility is controlled exclusively by OAI-SearchBot. Site owners should not attempt to use robots.txt to block ChatGPT-User as a mechanism for opting out of ChatGPT Search; use the OAI-SearchBot directive for that purpose instead.
What to do
Do not rely on blocking ChatGPT-User in robots.txt to control ChatGPT Search visibility - use OAI-SearchBot for that purpose. If you wish to be aware of ChatGPT-User traffic, monitor your server logs for the user-agent string: 'compatible; ChatGPT-User/1.0; +https://openai.com/bot'. Published IP ranges for ChatGPT-User are available at https://openai.com/chatgpt-user.json. No robots.txt change is required unless you have a specific policy reason to attempt to block user-initiated ChatGPT browsing.
Emerging agentic commerce protocols - the Agent Commerce Protocol (ACP) and Universal Commerce Protocol (UCP) - are being developed to enable AI agents to discover, evaluate, and transact with ecommerce sites programmatically. These protocols extend the agent-friendly-website guidance with specific structured interfaces that allow AI shopping agents to pull product information, pricing, availability, and checkout flows without relying solely on HTML parsing or visual rendering. Google has referenced UCP in its AI optimization guide as an emerging standard for Search agents. Botify research indicates that AI bots are significantly increasing their crawl of the web and that ecommerce sites need to prepare their product data for agent-driven discovery. While neither ACP nor UCP is yet a formal ranking signal, early adoption positions ecommerce sites for visibility in AI agent-driven shopping experiences as these protocols mature.
What to do
Review the Universal Commerce Protocol (UCP) documentation and the Agent Commerce Protocol (ACP) specification as they become available. For ecommerce sites: (1) ensure your product data is accessible via clean, crawlable HTML with complete structured data (Product, Offer, price, availability, shipping); (2) evaluate whether implementing a UCP or ACP endpoint is appropriate for your platform; (3) ensure your robots.txt does not block AI shopping agent crawlers; (4) review Botify's feed readiness checklist for AI discovery at https://www.botify.com/blog. Monitor Google Search Central and Bing Webmaster announcements for formal protocol support guidance.
IndexNow is an open protocol supported by Bing, Yandex, and other search engines (with Google evaluating adoption) that allows websites to instantly notify participating search engines when content is published, updated, or deleted. Rather than waiting for search engine crawlers to discover changes on their own schedule, IndexNow enables a push-based notification model that can significantly accelerate the time between content publication and indexing. This is particularly valuable for sites that publish or update content frequently (news, ecommerce product pages, job listings) and for mid-sized brands that may not have the crawl budget priority of large established sites. Botify has partnered with IndexNow and recommends it as a strategy for accelerating organic search growth. Bing Webmaster Tools provides native IndexNow integration.
What to do
Implement IndexNow by: (1) generating an IndexNow API key and hosting the key file at https://yourdomain.com/{key}.txt; (2) submitting URL change notifications to the IndexNow endpoint (https://api.indexnow.org/indexnow) whenever pages are published, updated, or deleted; (3) configuring your CMS or deployment pipeline to automatically trigger IndexNow submissions on content changes; (4) verifying implementation via Bing Webmaster Tools IndexNow report. Example submission: POST https://api.indexnow.org/indexnow with body
{
"host": "yourdomain.com",
"key": "your-key",
"urlList": [
"https://yourdomain.com/updated-page"
]
}
. Reference: https://www.indexnow.org/documentation
Google has refreshed the European Search Dataset Licensing Program page with updated program overview, eligibility criteria, and application details. Publishers in the European Economic Area (EEA) who wish to license their content to Google Search should review the updated eligibility criteria and application process. This program is distinct from standard indexing and is relevant only to publishers seeking a formal licensing arrangement with Google for their content in European Search.
What to do
Review the updated European Search Dataset Licensing Program page at https://developers.google.com/search/docs/appearance/european-search-dataset-licensing to determine eligibility and apply if appropriate for your organization. No code change is required; this is an administrative and contractual process.
Google Search supports Vacation Rental structured data, which allows vacation property listings to appear as rich results in Google Search. The structured data can include the property name, description, images, location, rating, and reviews. Sites that list vacation rental properties should implement this structured data to improve discoverability and eligibility for rich result appearances in Google Search.
What to do
Add VacationRental structured data to your vacation property listing pages: <script type="application/ld+json">{"@context":"https://schema.org","@type":"LodgingBusiness","name":"Property Name","description":"Property description","image":"https://yourdomain.com/property-image.jpg","address":{"@type":"PostalAddress","streetAddress":"123 Main St","addressLocality":"City","addressCountry":"US"},"aggregateRating":{"@type":"AggregateRating","ratingValue":"4.5","reviewCount":"100"}}</script>. Review Google's vacation rental structured data documentation for the full list of supported properties.
Google Search supports Movie structured data, which enables movie carousel rich results in Google Search (for example, 'best movies of 2023'). Sites that publish movie listing pages, movie databases, or editorial movie content can implement Movie structured data to provide details such as the movie title, director information, and images, making the content eligible for the movie carousel rich result experience.
What to do
Add Movie structured data to your movie listing or detail pages: <script type="application/ld+json">{"@context":"https://schema.org","@type":"Movie","name":"Movie Title","director":{"@type":"Person","name":"Director Name"},"image":"https://yourdomain.com/movie-poster.jpg","dateCreated":"2026-01-01","description":"Movie description"}</script>. Review Google's movie structured data documentation for the full list of supported properties and carousel eligibility requirements.
Google has adjusted its enforcement approach for the site reputation abuse policy within the European Economic Area (EEA). The site reputation abuse policy prohibits hosting third-party content on a site in an attempt to exploit the site's ranking signals, regardless of whether there is first-party involvement or oversight of the content. Site owners operating in the EEA should review the updated site reputation policy to understand how enforcement differs in that region and ensure their content hosting practices comply with both the global policy and the EEA-specific enforcement approach.
What to do
Review the updated site reputation abuse policy at https://developers.google.com/search/docs/essentials/spam-policies#site-reputation-abuse for EEA-specific enforcement guidance. Audit any third-party content hosted on your site to ensure it is not being used to exploit your site's ranking signals. If you operate in the EEA, review the specific enforcement adjustments that apply to your region.
Google has added a new guide on how to analyze social and video platform content performance in Search Console. Content creators, social media managers, and SEO professionals can now understand how their social and video platform content (e.g., YouTube videos, social posts) performs on Google Search through dedicated Search Console reporting. Site owners with social or video platform content should review this guide to understand how to measure and optimize their content's search performance across platforms.
What to do
Review the new Search Console guide on social and video platform content performance. Connect your social and video platform accounts to Search Console where supported. Monitor performance data for your platform content in Search Console Performance reports to understand how it appears in Google Search results.
Google has added a new guideline to the review snippet documentation specifically addressing fake and undisclosed incentivized reviews. Sites that collect or display fake reviews, or reviews where an incentive was provided but not disclosed, risk losing eligibility for review rich results and may be subject to manual action under Google's spam policies. This applies to both first-party reviews collected on your site and third-party review markup you implement. All incentivized reviews must be clearly disclosed as such. Fake reviews of any kind are prohibited.
What to do
Audit your review collection and display practices: (1) remove any fake or fabricated reviews from your structured data and visible content; (2) ensure any reviews where an incentive was provided (discount, free product, payment) are clearly labeled as incentivized; (3) update your review schema to only include genuine, unbiased reviews or properly disclosed incentivized reviews; (4) implement review policies that prohibit fake reviews and require disclosure of incentives. Do not implement ReviewSnippet structured data for reviews that violate these guidelines.
Google has updated the package tracking documentation to reflect that the package tracking early adopters program is no longer accepting new partners. Sites that were not already accepted into the program cannot apply to participate. Existing partners should review the updated feature availability and eligibility requirements sections of the package tracking documentation to ensure continued compliance.
What to do
If you are an existing package tracking early adopter partner, review the updated feature availability and eligibility requirements in the package tracking documentation to ensure your implementation remains compliant. If you are not yet a partner, the program is closed to new applicants - monitor Google Search Central for any future announcements about program expansion.
Google has updated the canonicalization troubleshooting guide with clarifications on re-evaluation time. When you make changes to canonical tags, hreflang annotations, or other canonicalization signals, Google does not immediately re-evaluate and update its canonical selection. Site owners should set realistic expectations: canonicalization changes can take days to weeks to be reflected in Google Search, depending on crawl frequency and site size. Do not make rapid successive changes to canonical signals while waiting for previous changes to take effect, as this can create conflicting signals and further delay resolution.
What to do
After making canonicalization changes (updating canonical tags, fixing hreflang, resolving duplicate content): (1) use Google Search Console URL Inspection to request indexing of affected pages; (2) allow sufficient time (days to weeks) for Google to re-crawl and re-evaluate canonical signals before concluding a change has not worked; (3) monitor Search Console Coverage and URL Inspection reports to track canonicalization status; (4) avoid making additional conflicting canonicalization changes while waiting for previous changes to propagate.
Google has updated the product structured data documentation to detail how the google_product_category property can be used with both Product and Offer types. This aligns with the Google Merchant Center feed specification for the google_product_category attribute. Providing both merchant-defined category information and Google-defined category information within schema.org markup enhances product information for Google Search and Shopping, improving product discoverability and classification accuracy in merchant listings.
What to do
Add category information to your Product or Offer structured data using the appropriate property for google_product_category. Include both your merchant-defined category and the Google product taxonomy category where applicable:
{
"@type": "Product",
"name": "Product Name",
"category": "Your Category > Subcategory"
}
. Review the Google Merchant Center product category taxonomy and the updated product structured data documentation for the specific property and value format required.
Google has added a new section on 'Sale duration' to the Merchant listing guide explaining how to use schema.org properties to set the effective range for sale prices. This aligns schema.org usage with the Merchant Center feed attribute sale_price_effective_date. Specifying sale price effective dates helps Google accurately display sale pricing information in merchant listings and product rich results, including the duration of the sale. Without effective date information, Google cannot determine when a sale price is valid, which may result in incorrect or missing sale price display.
What to do
Add sale price effective date information to your Offer structured data using the priceValidUntil property and, where applicable, the priceValidFrom equivalent. Example:
{
"@type": "Offer",
"price": "19.99",
"priceCurrency": "USD",
"priceValidUntil": "2026-12-31"
}
. For sale prices with both start and end dates, review the updated Merchant listing guide for the recommended schema.org property combination. Ensure your implementation aligns with the sale_price_effective_date attribute format used in Merchant Center feeds.
Google Search has updated how it connects users to AMP pages. Starting with this change, Google will now take users directly to the publisher's AMP host pages rather than serving them from the AMP Cache or via signed exchanges. This simplifies and reduces maintenance efforts for publishers creating AMP content. Publishers no longer need to update the AMP Cache or configure signed exchanges for their AMP content. AMP content will continue to rank just like any other web page. Existing AMP Cache configurations and signed exchange setups will not cause harm but are no longer required for AMP pages to appear in Google Search.
What to do
No immediate action required. If you maintain AMP pages: (1) you no longer need to configure or maintain signed exchanges (SXG) specifically for AMP; (2) you no longer need to manually update the AMP Cache; (3) ensure your AMP pages are accessible at their canonical publisher URLs as Google will now link directly to those URLs; (4) review and update any internal documentation that references AMP Cache or signed exchange requirements for AMP. AMP content continues to be supported and will rank normally.
Google has released an August 2026 Spam Update. Spam updates target sites that violate Google's spam policies, including cloaking, doorway pages, scaled content abuse, site reputation abuse, back button hijacking, fake reviews, and other prohibited practices. Sites affected by spam updates may experience significant ranking drops or removal from Google Search results. Recovery from a spam update requires identifying and resolving the spam policy violation, then submitting a reconsideration request if a manual action was applied.
What to do
Audit your site against Google's full spam policies at https://developers.google.com/search/docs/essentials/spam-policies. Check Google Search Console for any manual action notifications. If your site experienced a traffic drop around August 2026, review your content and practices against all spam policies including: cloaking, doorway pages, scaled content abuse, site reputation abuse, back button hijacking, fake or undisclosed incentivized reviews, and link spam. Resolve any violations and submit a reconsideration request if a manual action was applied.
Schema.org version 30.0 added GS1, Dublin Core, and Open Graph equivalence annotations and exports to the schema.org vocabulary. This means schema.org properties now have formally declared equivalences to terms in these widely-used external vocabularies. For ecommerce sites using GS1 identifiers, publishers using Dublin Core metadata, and sites using Open Graph markup, these equivalence annotations improve interoperability between schema.org structured data and other metadata systems. Sites that implement both schema.org structured data and Open Graph tags can now benefit from formally declared cross-vocabulary alignment. Additionally, version 30.0 extended credential vocabulary to allow specification of credentials beyond just educational credentials, and extended CompoundPriceSpecification to have the more generic UnitPriceSpecification.
What to do
No immediate code change required. If your site uses GS1 product identifiers, Dublin Core metadata, or Open Graph tags alongside schema.org structured data, review the schema.org version 30.0 release notes to understand the new equivalence mappings. Ensure your structured data and Open Graph tags are consistent with each other, as the formal equivalence annotations may affect how search engines and other consumers interpret cross-vocabulary signals.
Schema.org version 30.0 added EU Digital Product Passport (DPP) examples aligned with UN/CEFACT codes. The EU Digital Product Passport is a regulatory requirement under the EU Ecodesign for Sustainable Products Regulation (ESPR) that will require products sold in the EU to carry a digital record of sustainability, repairability, and lifecycle information. Schema.org now provides vocabulary and examples for implementing DPP data in structured data. Ecommerce sites targeting EU markets should be aware of this emerging structured data vocabulary as DPP requirements come into force.
What to do
Review the schema.org version 30.0 release notes and EU Digital Product Passport examples at https://schema.org/docs/releases.html. If your site sells products in EU markets, evaluate whether implementing DPP-aligned structured data is relevant to your compliance and SEO strategy. The vocabulary is aligned with UN/CEFACT codes for interoperability with regulatory and supply chain systems.
Schema.org version 29.0 added vocabulary to represent financial incentives, for example in the area of sustainability. New types and enumeration types (including IncentiveQualifiedExpenseType) were added, along with new properties for use on Offer including eligibleWithSupplier. This vocabulary is particularly relevant for ecommerce sites that offer sustainability-related incentives, rebates, or government-backed financial incentives on products. Implementing financial incentive structured data can help search engines accurately represent offer conditions in merchant listings and product rich results.
What to do
If your site offers products with associated financial incentives (e.g., sustainability rebates, government incentives, supplier-specific pricing), review the schema.org version 29.0 financial incentives vocabulary at https://schema.org/docs/releases.html. Add the appropriate incentive properties to your Offer structured data. Example:
{
"@type": "Offer",
"price": "299.99",
"priceCurrency": "USD",
"eligibleWithSupplier": {
"@type": "Organization",
"name": "Supplier Name"
}
}
. Review the full vocabulary for IncentiveQualifiedExpenseType and related enumeration values.
Google has indicated that Google Search will eventually support the HTTP QUERY method, an emerging web standard for safe, cacheable requests with a request body (unlike GET which has no body, and unlike POST which is not safe or cacheable). The HTTP QUERY method is being standardized to enable more expressive search and query interactions on the web. No immediate implementation is required, but site owners and developers building search interfaces or APIs should be aware of this emerging standard and monitor its adoption timeline.
What to do
No immediate action required. Monitor the HTTP QUERY method standardization progress and Google Search's implementation timeline. If you build search APIs or query interfaces, evaluate whether adopting the HTTP QUERY method when it becomes available would benefit your architecture. Do not implement HTTP QUERY for SEO purposes until Google formally announces support and provides implementation guidance.
Google has updated its site move guidance to explicitly include information on using the Change of Address tool for all subdomain variants (including www and non-www) during domain migrations. When performing a domain migration, domain moves work best when all variants of a site are migrated properly. Failing to migrate all subdomain variants (e.g., migrating example.com but not www.example.com, or vice versa) can result in incomplete migration signals and slower consolidation of ranking signals to the new domain.
What to do
When performing a domain migration, use the Change of Address tool in Google Search Console for all subdomain variants of your site, including both www and non-www versions. Ensure 301 redirects are in place from all variants of the old domain to the corresponding variants of the new domain. Verify each variant is submitted separately in Search Console if they are tracked as separate properties.
Google has updated the preferred sources feature to now be available in AI Overviews and AI Mode, in addition to standard Google Search. Publishers who have configured preferred sources can now benefit from increased visibility in AI-generated search experiences. To take advantage of this, sites must meet the eligibility requirements for the preferred sources feature and ensure their identity signals (Organization schema, site name schema, canonical URLs) are correctly implemented.
What to do
Review Google's preferred sources documentation to understand eligibility and configuration requirements for AI Overviews and AI Mode. Ensure your site's identity signals are correctly implemented: Organization schema, WebSite schema with site name, and consistent canonical URLs. Download and use the updated translated button assets if applicable.
Google has added support for the hasAdultConsideration property in merchant listing structured data, bringing parity with the Merchant Center feed specification. This property should be used on product or offer pages that contain adult content to help Google correctly classify and filter the content in shopping experiences. Without this property, Google's algorithms must attempt to automatically detect adult content in merchant listings, which may result in incorrect classification.
What to do
Add the hasAdultConsideration property to your Product or Offer structured data for adult content listings:
{"@type":"Product","name":"Product Name","hasAdultConsideration":true, ...}
. This brings your structured data markup into parity with the Merchant Center feed specification for adult products.
Google has published new guidance on evaluating third-party SEO tools, services, and advice, including important considerations when assessing recommendations from SEO tools and consultants. Key points include: (1) not all SEO tool recommendations reflect Google's actual ranking signals or guidelines; (2) site owners should cross-reference tool recommendations against official Google Search Central documentation; (3) some SEO advice may be outdated, incorrect, or based on correlation rather than causation; (4) Google's own documentation and Search Console data should be the primary reference for SEO decisions. The guidance has been added to the page on evaluating SEO recommendations and tools.
What to do
When evaluating SEO tool recommendations: (1) cross-reference all recommendations against official Google Search Central documentation at https://developers.google.com/search/docs; (2) use Google Search Console as the primary source of truth for your site's performance data; (3) be skeptical of recommendations that cannot be traced to official Google guidance; (4) prioritize fixes flagged by Google Search Console over third-party tool warnings when they conflict.
Google has removed outdated guidance stating that using JavaScript to load content makes it 'harder for Google Search' or harder for assistive technologies. Google Search has been rendering JavaScript for multiple years and most assistive technologies now work with JavaScript. Site owners and SEO tools should not flag JavaScript-rendered content as an inherent SEO or accessibility problem. The concern about JavaScript content being invisible to Googlebot is no longer valid for standard client-side rendering scenarios, though server-side rendering remains a best practice for performance and reliability.
What to do
No action required if you are using JavaScript to load content. Standard client-side rendering is supported by Google Search. If you have internal documentation or tooling that flags JavaScript content loading as an SEO issue, update it to reflect current Google guidance. Server-side rendering (SSR) or static rendering is still recommended for performance and Core Web Vitals reasons, but is not required for indexing.
Google has migrated its core crawling documentation to a new dedicated site at Google's crawling infrastructure documentation. The migrated documentation includes: overview of Google crawlers and fetchers, verifying requests from Google crawlers, reducing Google crawl rate, Google common crawlers, how Google interprets the robots.txt specification, managing crawling of faceted navigation URLs, optimizing crawl budget, how HTTP status codes affect crawlers, and debugging DNS and network errors. The functionality and guidance has not changed - only the location of the documentation. A separate changelog feed is now available for crawling infrastructure updates at https://developers.google.com/crawling/docs/changelog/crawling_docs_updates.rss. This migration reflects that Google's crawling infrastructure is shared across multiple Google products beyond Search, including Google Shopping, News, Gemini, AdSense, and more.
What to do
Update any internal documentation, bookmarks, or tooling that references the old crawling documentation URLs to point to the new Google crawling infrastructure site. Subscribe to the new crawling infrastructure changelog feed at https://developers.google.com/crawling/docs/changelog/crawling_docs_updates.rss to receive updates about crawling behavior changes that may affect your site across all Google products, not just Search.
Google's structured data carousels (beta) feature has been expanded to Turkey, in addition to South Africa and other previously supported regions. If you operate a site in Turkey with summary pages that link out to detail pages (such as travel, local, shopping, or other supported query types), implementing carousel structured data markup may make your content eligible for carousel rich results in Google Search in Turkey. The markup must be placed on the summary page; you do not need to add markup to the detail pages.
What to do
If your site operates in Turkey and has summary pages that link to detail pages in supported verticals (travel, local, shopping, and others), review Google's structured data carousels (beta) documentation to implement the appropriate carousel markup. Place the markup on the summary page only. Use the interest form in the documentation to express interest in the feature.
Google has clarified that spam report submissions may now be used to take manual action against sites that violate Google's spam policies. Previously, spam reports were used primarily to improve Google's automated systems and were not directly used to trigger manual actions. Site owners should be aware that user-submitted spam reports can now result in manual penalties. This makes compliance with Google's spam policies more important than ever, as violations are now more likely to result in direct manual action in addition to algorithmic penalties.
What to do
Audit your site against Google's full spam policies at https://developers.google.com/search/docs/essentials/spam-policies. Ensure your site does not engage in cloaking, doorway pages, scaled content abuse, site reputation abuse, back button hijacking, or any other prohibited practices. Monitor Google Search Console for any manual action notifications. If you receive a manual action, follow the reconsideration request process after resolving the violation.
Schema.org version 24.0 introduced the IPTCDigitalSourceEnumeration vocabulary with 14 coded values to describe the digital source type of media, including AI-generated and algorithmically manipulated content (e.g., TrainedAlgorithmicMediaDigitalSource, CompositeWithTrainedAlgorithmicMediaDigitalSource). Schema.org version 29.4 added additional missing enumeration values to IPTCDigitalSourceEnumeration. This vocabulary is used via the digitalSourceType property on MediaObject and its subtypes (ImageObject, VideoObject, AudioObject). Correctly tagging AI-generated or AI-assisted media with the appropriate IPTC digital source type supports content provenance signals used by Google's 'About this image' feature and aligns with C2PA metadata standards. This rule complements the c2pa-metadata rule.
What to do
Add the digitalSourceType property to your ImageObject or VideoObject structured data using the appropriate IPTCDigitalSourceEnumeration value. For AI-generated images:
{
"@type": "ImageObject",
"contentUrl": "https://yourdomain.com/image.jpg",
"digitalSourceType": "https://cv.iptc.org/newscodes/digitalsourcetype/trainedAlgorithmicMedia"
}
. For AI-composite images use: "https://cv.iptc.org/newscodes/digitalsourcetype/compositeWithTrainedAlgorithmicMedia". For human-captured original content use: "https://cv.iptc.org/newscodes/digitalsourcetype/originalDigitalCapture".
Bing has launched 'AI Performance' in Bing Webmaster Tools (now in public preview), which shows when your site's URLs are cited in AI-generated answers across Microsoft Copilot, AI-generated summaries in Bing, and select partner integrations. To be eligible for citation in Bing AI-generated answers, pages must be indexed by Bing and eligible to appear in Bing Search with a snippet. Pages blocked by robots.txt, carrying noindex directives, or carrying nosnippet directives will not be cited in Bing AI features. Site owners should monitor citation activity and URL-level citation data via Bing Webmaster Tools to understand their visibility in AI-generated answers. Standard Bing webmaster best practices - clean crawlable HTML, descriptive titles and meta descriptions, structured data, and fast page load - remain the foundation for AI citation eligibility.
What to do
Verify your pages meet Bing's standard indexing and snippet eligibility requirements: (1) confirm pages are not blocked by robots.txt; (2) confirm no noindex or nosnippet meta tags or X-Robots-Tag headers are present; (3) ensure pages return 200 HTTP status codes; (4) submit your sitemap in Bing Webmaster Tools; (5) monitor AI citation performance via the AI Performance report in Bing Webmaster Tools at https://www.bing.com/webmasters.