Google Search Guidelines: The Complete Cheatsheet

TL;DR

• Make pages crawlable, indexable, and useful, but don’t expect Google to guarantee rankings or indexing.
• Avoid spam tactics such as cloaking, keyword stuffing, link spam, and scaled low-value content.
• Create original, trustworthy content for people; Google says trust matters most.
• AI Overviews need no special files, markup, or writing tricks—make pages eligible for Search.
• Use Search Console to track indexing, site issues, and traffic changes.

How Google Search Works: Crawling, Indexing and Serving Results

Follow Google’s three-stage process: make pages discoverable, indexable, and useful for relevant queries; rules checked against Google Search Central in October 2026.

• Understand the stages: crawling discovers and downloads pages; indexing analyzes and stores them, choosing a canonical page among duplicates; serving returns results relevant to the query. Google says, “Google doesn't guarantee that it will crawl, index, or serve your page, even if your page follows the Google Search Essentials.” It also says, “Google doesn't accept payment to crawl a site more frequently, or rank it higher.” Source: how Google Search works.
• Let Googlebot render your JavaScript; Google renders pages using a recent version of Chrome.
• Don’t pay anyone who promises faster Google indexing or better Google rankings; Google does not sell either outcome.

Google Search Essentials: The 3 Technical Requirements and Key Best Practices

Use Search Essentials as the baseline: its three parts are technical requirements, spam policies, and key best practices. Source: Google Search Essentials.

• Let Googlebot access the page; don’t block it from crawling. Source: technical requirements.
• Make the page work and return an HTTP 200 status code to Google.
• Give the page indexable content. Meeting all three minimum technical requirements does not guarantee that Google will index a page.
• Follow Google’s spam policies; don’t use tactics that violate them.
• Create helpful, reliable, people-first content.
• Put the words people search for in prominent places, including titles, headings, alt text, and link text.
• Make links crawlable so Google can discover pages.
• Promote your site in relevant communities.
• Follow Google’s best practices for images, video, structured data, and JavaScript.
• Enable the search features that fit your site.
• Use Google’s official controls to manage how your content appears in Search.

Google Spam Policies: The 16 Practices That Can Get a Site Demoted or Removed

Follow Google’s spam policies to avoid lower rankings, removal from Search results, or a manual action. Source: Google's spam policies.

• Don’t use cloaking.
• Don’t engage in doorway abuse.
• Don’t engage in expired domain abuse.
• Don’t publish hacked content.
• Don’t use hidden text and link abuse.
• Don’t engage in keyword stuffing.
• Don’t engage in link spam.
• Don’t generate machine-generated traffic.
• Don’t use malicious practices.
• Don’t use misleading functionality.
• Don’t engage in scaled content abuse: generating many pages mainly to manipulate rankings, “no matter how it's created.”
• Don’t scrape content.
• Don’t break the site reputation policy: don’t publish third-party content mainly to exploit the host site’s ranking signals.
• Don’t use sneaky redirects.
• Don’t engage in thin affiliation.
• Don’t allow user-generated spam.

If Search Console reports a manual action, fix the issue and then file a reconsideration request in Search Console.

What Google Means by Helpful, Reliable, People-First Content

Google favors content created to help people rather than to win search visits.

• Do show who made the content: add a byline where readers expect one and explain the author’s relevant background. Source: helpful, reliable, people-first content.
• Do explain how you made the content when that context would help readers understand it. If automation or AI substantially generated it, disclose that use where readers might reasonably ask how the content was created.
• Do make the reason for publishing clear: create the page to help its intended audience, not primarily to manipulate search rankings.
• Don’t use automation primarily to manipulate rankings; Google says this violates its spam policies.
• Do build E-E-A-T: experience, expertise, authoritativeness, and trustworthiness. Google says, “Of these aspects, trust is most important.”
• Do apply extra care to health, financial, and safety content: Google gives stronger weight to signals related to trust for topics that could significantly affect people’s well-being.
• Don’t produce content mainly to attract search visits instead of helping readers.
• Don’t use extensive automation to produce content across many topics without adding value for people.
• Don’t publish work “mainly summarizing what others have to say without adding much value”.
• Don’t write about a trending topic just because it may attract visits if it is not relevant to your audience or expertise.
• Do assess each page for original information, clear sourcing, first-hand expertise, and a title that describes the content without exaggerating it.

How Google AI Overviews and AI Mode Choose Pages to Link

Treat AI Overviews and AI Mode as Search features: make useful pages eligible for Search, not specially optimized for AI. Source: AI features and your website.

• Don’t build an “AI hack”: Google says, “There are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary.”
• Expect query fan-out: both features may issue multiple related searches across subtopics and data sources to build an answer.
• Make each supporting page indexed and eligible to appear in Google Search with a snippet.
• Allow Google’s crawlers in robots.txt and in your CDN or hosting settings.
• Link important pages from other pages on your site so Google can find them.
• Keep important information in text, and support it with high-quality images and video.
• Keep structured data consistent with the visible content on the page.
• Keep your Merchant Center and Business Profile information accurate and up to date.
• Create unique content rather than commodity summaries; give readers first-hand insight, such as a review based on actual experience, instead of restating information available elsewhere.

What Google Says You Don't Need for AI Search: llms.txt, Chunking, AI Schema

For Google Search, you can ignore tactics like chunking content, creating unnecessary AI text files (like llms.txt), or pursuing inauthentic mentions. These rules apply to Google Search only, not other AI assistants. Source: optimizing for generative AI search.

• Don’t create llms.txt or other AI text files for Google Search: Google says, “Google Search itself doesn't use them.” Creating a file for other services “will neither harm nor help” visibility in Google Search.
• Don’t break content into tiny chunks for AI: Google says, “There's no requirement to break your content into tiny pieces for AI to better understand it.”
• Don’t adopt a special AI writing style: Google says, “You don't need to write in a specific way just for generative AI search.”
• Don’t chase inauthentic mentions to influence AI results; Google lists them among the tactics to ignore.
• Don’t add special schema for generative AI search: Google says, “there's no special schema.org markup you need to add.” Structured data can still help pages qualify for rich results.
• Don’t rely on third-party tools’ claims about Google’s internal systems: Google warns, “No third-party tool has access to our internal ranking or AI systems.”

How to Control How Google Uses Your Content in AI Overviews and AI Mode

Use Search preview controls to manage content in AI Overviews and AI Mode.

• Use nosnippet to prevent Google from showing a snippet, or data-nosnippet to hide selected page content from snippets. Source: robots meta tag, data-nosnippet and X-Robots-Tag.
• Set max-snippet to cap the length of text Google may show in a snippet.
• Use noindex to remove a page from Google Search; pages removed from Search can’t appear in AI Overviews or AI Mode.
• Choose nosnippet or noindex if you want to prevent a page from appearing in AI Overviews or AI Mode; blocking Google-Extended alone does not do this.
• Don’t treat Google-Extended as a Search visibility control: it governs whether content Google crawls may be used to train future Gemini models or ground responses in Gemini Apps and Vertex AI. Google says, “Google-Extended does not impact a site's inclusion in Google Search nor is it used as a ranking signal in Google Search.” Source: Google's common crawlers.
• Check whether Google sees a control with URL Inspection, then request a recrawl if you changed the page or directive.

Google Crawling and Indexing Checklist

Use crawl controls to guide discovery and indexing without sending conflicting signals.

• Do use robots.txt to manage crawling, not to keep a page out of Google; Google says it is “not a mechanism for keeping a web page out of Google.” Source: introduction to robots.txt.
• Do use noindex when you want a page excluded from Google, and leave it crawlable so Google can see the directive. For PDFs and other non-HTML files, send noindex in an X-Robots-Tag HTTP header. Source: block indexing with noindex.
• Do keep each sitemap to no more than 50,000 URLs or 50MB uncompressed; use absolute URLs, UTF-8, and canonical URLs only. Source: build and submit a sitemap.
• Don’t rely on sitemap <priority> or <changefreq> values; Google ignores them. Use <lastmod> only when it consistently reflects the page’s accurate last-modified date.
• Do submit your sitemap in Search Console or include a Sitemap: line in robots.txt.
• Do consolidate duplicate URLs with redirects or rel=canonical; treat sitemap inclusion as a weaker canonical signal. Source: specify a canonical URL.
• Don’t use robots.txt, noindex, or the URL removal tool to choose a canonical URL, and don’t declare conflicting canonicals.
• Do use <a href> elements for links, write descriptive anchor text instead of “click here,” and link every important page from at least one other page. Source: crawlable links.
• Do mark paid links with rel=sponsored and user-generated links with rel=ugc; use nofollow when you don’t trust the target.
• Do use 301 or 308 redirects for permanent moves and 302 or 307 redirects for temporary ones; Google treats the destination of a permanent redirect as canonical. Source: redirects and Google Search.
• Do implement redirects server-side first; use JavaScript redirects only when server-side options aren’t possible.
• Don’t change a page’s canonical URL with JavaScript or rely on JavaScript to remove a noindex; Google may skip rendering. Source: JavaScript SEO basics.
• Do keep the same content, structured data, meta robots directives, titles, and descriptions on mobile and desktop. Source: mobile-first indexing.
• Don’t make primary content load only after a user interacts with the page; Google recommends responsive design.
• Do make URLs readable, separate words with hyphens rather than underscores, and don’t use fragments (#) to change page content. Source: URL structure.

How Google Builds Title Links, Snippets and Rich Results

Give each page a clear title link, accurate snippet, and eligible structured data.

• Write a unique, descriptive, concise <title> for every page; avoid keyword stuffing and vague labels such as “Home.” Source: title links.
• Put the page’s main title in a prominent first <h1>, and include the site name once in the title, separated by a delimiter.
• Keep titles useful rather than targeting a character count: Google may build the title link from the <title>, headings, og:title, anchor text, or WebSite structured data, and may truncate it to fit the device width.
• Write page-specific meta descriptions with facts that describe that page; there is no length limit, and Google may use the page content instead when it better fits the query. Source: snippets and meta descriptions.
• Expect snippets to come mainly from page content; write a meta description that summarizes the page accurately rather than assuming Google will show it.
• Add WebSite structured data to the home page with the preferred site name and, optionally, an alternate name. Source: site names.
• Use JSON-LD for structured data where practical; mark up only content readers can see, and don’t use markup to mislead. Source: structured data general guidelines.
• Validate structured data with Google’s Rich Results Test, and don’t promise a rich result: eligibility does not guarantee display. Policy violations can trigger a manual action that removes rich result eligibility.
• Use a square favicon at least 8x8px; Google recommends a size larger than 48x48px. Keep its URL stable and make the home page crawlable. Source: favicon guidelines.

Google Core Web Vitals Thresholds and Page Experience Checklist

Meet Google’s Core Web Vitals “good” thresholds and check the page experience basics.

• Keep LCP within 2.5 seconds. Source: Core Web Vitals.
• Keep INP under 200 milliseconds.
• Keep CLS under 0.1.
• Measure Core Web Vitals in the Search Console Core Web Vitals report.
• Treat page experience as a set of signals, not a single ranking factor: Google says, “There is no single signal. Our core ranking systems look at a variety of signals that align with overall page experience.” Source: page experience.
• Use HTTPS.
• Make content display well on mobile.
• Don’t let excessive ads distract from the main content.
• Don’t use intrusive interstitials.
• Make the main content easy to distinguish from other page content.

Google Ranking Systems and Core Updates: What to Do After a Traffic Drop

Check whether a core update explains the traffic change before changing pages.

• Read Google’s ranking systems guide to understand the systems it names: BERT, freshness, link analysis and PageRank, neural matching, original content, passage ranking, RankBrain, reliable information, reviews, site diversity, and spam detection. Source: Google's ranking systems guide.
• Treat the helpful content system as part of Google’s core ranking systems; it is no longer a separate system.
• Check the Search Status Dashboard and wait until the core update rollout has finished before assessing its impact. Source: core updates.
• Wait “at least a full week” after the rollout completes, then compare Search Console performance for a week after the update with a week before it.
• Don’t rush to revise pages after a small position change, such as position 2 to 4; Google says this usually does not call for action.
• Review the site as a whole against Google’s helpful content questions if the drop is large; don’t assume one page or one ranking signal explains it.
• Allow time for recovery after making improvements: Google says it can take from a few days to “several months” for its systems to recognize and reflect changes.
• Use Google’s guidance on debugging drops in Search traffic to check whether the decline is limited to Search or reflects a broader traffic change (see the Search Console section below).

Google Rules for Multilingual and Paginated Websites

Give each language and paginated page a crawlable, distinct URL, and annotate alternate versions correctly.

• Do give each language version its own URL; don’t switch language based on cookies or a user’s IP address. Source: multi-regional and multilingual sites.
• Don’t rely on the lang attribute to tell Google a page’s language: Google detects language from the visible content.
• Do translate the full page, including navigation and supporting content, rather than translating only the main text.
• Don’t automatically redirect visitors based on their perceived language; offer links so they can choose a version.
• Do add hreflang annotations in the HTML <head>, an HTTP header, or a sitemap to identify localized versions. Source: localized versions with hreflang.
• Do make every version list itself and all alternate versions; use reciprocal annotations or Google may ignore them.
• Do use fully qualified URLs and an ISO 639-1 language code, with an optional ISO 3166-1 region code such as en-GB; don’t use a region alone. UK and EU are not valid region codes.
• Do use x-default to identify the fallback page for users whose language or region doesn’t match another version.
• Do link paginated pages in sequence using <a href> links, and give each page its own URL and canonical. Source: pagination.
• Don’t canonicalize every page in a series to page one or use URL fragments (#) to distinguish pages.
• Don’t use rel=next or rel=prev to signal pagination; Google no longer uses these link relations.

How to Monitor Google Search Performance With Search Console

Use Search Console to monitor indexing, performance, and site issues.

• Verify your site in Search Console and submit a sitemap. Source: get started with Search Console.
• Don’t check Search Console every day: Google emails you about new issues and suggests checking around once a month or after content changes.
• Review the Performance, Page indexing, Core Web Vitals, Manual actions, and Security issues reports.
• Use URL Inspection to test a live URL and request a recrawl.
• Check the Generative AI performance report in Search Console to monitor performance in Google’s AI features.
• When traffic drops, investigate Google’s six listed causes: algorithmic updates, technical issues, security issues, spam violations, seasonality and changing interest, and site migrations. Source: debugging drops in Search traffic.
• Compare equivalent periods and use “Last 16 months” to look for seasonal patterns.
• Check whether the decline affects the whole site or only some pages, then sort pages by click difference to find the largest losses.
• Cross-check changes in search interest with Google Trends.

Google Search Guidelines Quick-Reference Table

Use this table as a one-line summary of each section; every rule comes from the official Google page named in the last column.