Signal to BuildSIGNAL TO BUILD中文Owner sign in ↗
← Knowledge Notes
Global Growth ／ Knowledge note · English

Complete Visual Elements of Search Results (Visual Gallery / Title / Snippet / Favicon / Sitelinks / Featured / AI / Translated / Preferred)

All visual elements on Google Search result pages. 4 result types (Text/Rich/Image/Video) plus Exploration features. Attribution (Favicon/Site name/Visible URL/Domain/Breadcrumb). **Title link**: 11 best practices + 6 major Google override scenarios (half-empty/obsolete/inaccurate/duplicate/no clear main content/lang mismatch) + 9 sources. **Snippet** elements + 6 meta description guidelines + 3 rules for Read more deep links. **Favicon**: 6 implementation rules (domain/subdomain level only + 8x8px minimum + must be crawlable by Googlebot-Image). **Sitelinks**: 4 guidelines + removal methods. **Featured Snippets**: fully blocked by nosnippet / partially blocked by max-snippet / cannot be actively marked up. **Enriched search** (Jobs/Recipes/Events are the only ones) + 5 quality guidelines. **AI Features** (AI Overviews + AI Mode): no special SEO requirements, **no AI text file / no special schema**, control with nosnippet/max-snippet/noindex + use Google-Extended to control AI training. **Translated results**: 21 languages + notranslate opt-out + 9-step algorithm for providing URL decoding to ad networks. **Preferred sources** (Top Stories): users actively select preferences + deeplinks + button assets.

Source collection：Google Search and SEO · Published here：2026-09-26

SEOSearch EnginesOrganic growth

In one sentence: The Google Search results page is an assembly of visual elements: every result type and every attribution element has its own control method. This wiki is a lookup table for "I want part X of the search result to look like this".

TL;DR — Which part of the results page do you want to control?

Element How to control
Title link Write good <title> + main visual title + heading(Part 2)
Snippet Write good meta descriptions + data-nosnippet exclude some parts (Part 3)
Favicon <link rel="icon"> + 8x8 min(Part 4)
Site name Mark up with WebSite SD (see Google SEO Basics + How Search Works + Search Essentials)
Breadcrumb BreadcrumbList SD (see Complete Ecommerce SEO Guide (6 Surfaces / 4 Launch Strategies / Merchant Center / 6 SD Types / 3 Pagination Modes / 13 Review Guidelines))
Sitelinks Cannot be directly controlled — good internal link structure + avoid duplication (Part 5)
Featured snippet Cannot be actively marked up — opt out with nosnippet / max-snippet (Part 6)
Rich attributes(stars/recipe/etc) structured data (see Complete Structured Data Overview (Framework + Quick Lookup for 34 Rich Result Types + JSON-LD/Microdata/RDFa Selection) to follow)
Enriched search(jobs/recipes/events) Corresponding SD + completeness (Part 7)
AI Overviews / AI Mode Standard SEO + nosnippet not displayed (Part 8)
Translated results notranslate opt out(Part 9)
Preferred sources Guide users to select you as a preferred source (Part 10)
Image / Video result google-images / video best practices (see Complete Guide to Rich Features / Discover / Images / Video / Page Experience / Web Stories / SXG / Paywall to follow)

Part 1 — Visual Elements Gallery: Full Panorama of Results Page Elements

1.1 The 5 major result types

Type When it appears Primary driver
Text result(plain blue link) Default Content quality + SEO
Rich result Triggered by structured data SD markup
Image result image-seeking query Image SEO
Video result video-seeking query Video SEO
Exploration features(People also ask / Related searches) Automatic Cannot be directly controlled

1.2 Attribution elements

Element Meaning
Favicon Site icon (Part 4)
Site name Site name (WebSite SD)
Visible URL Domain + breadcrumb
Domain example.com
Breadcrumb Hierarchy (BreadcrumbList SD)

1.3 Text result sub-elements

  • Attribution
  • Title link
  • Snippet
  • Byline date(publication-dates SD)
  • Sitelinks group(2+ sitelinks)
  • Text result image(most relevant in-page image)
  • Rich attributes(stars / recipe time / etc. — SD-driven)

1.4 Image / Video result sub-elements

  • Image thumbnail / Video thumbnail
  • Attribution
  • Upload date(video only)

1.5 Exploration features

  • Related searches group: Related terms other people search for
  • Related questions group(People Also Ask): expanding shows a featured snippet

⚠️ exploration features Cannot be directly controlled, but it is great inspiration for content topic selection.

Part 2 — Title Link

Roughly in descending order of priority:

  1. <title> element
  2. Main visual title(the most prominent title on the page)
  3. Heading elements (especially <h1>)
  4. og:title meta tag
  5. Other large-font / prominent text
  6. Other text on the page
  7. Anchor text on the page
  8. Text in links from other sites pointing to this page
  9. WebSite structured data

2.2 11 best practice

  1. Every page must have <title> — it is required
  2. descriptive + concise — don't use "Home" / "Profile".There is no word count limit but it is truncated to device width on the SERP
  3. Do not keyword stuff("Foobar, foo bar, foobars, foo bars")
  4. Do not use boilerplate / duplicates — each page must be unique; do not use "Cheap products for sale" across the whole site
  5. Keep the brand concise — it is fine to add the brand on the homepage, for inner pages use Site Name | Page Topic this format, do not repeat the full name on every page
  6. Make the main title visually clear — Google gets confused when multiple headings are at the same level
  7. Do not block important pages with robots.txt — after blocking, Google uses anchor text from other sites as the title
  8. Match the page's main language + writing system — a Hindi page should have a Hindi title (do not use Latin transliteration)
  9. Do not stuff in flight prices(they change by the second, Google will ignore them)
  10. After adjusting <title> it takes several days to several weeks before Google sees it — you can use URL Inspection to request a recrawl
  11. Google cannot manually change titles for individual sites — you can only change it yourself

2.3 The 6 major Google override scenarios

Scenario Example Google behavior
Half-empty <title> <title>\| Site Name</title>(empty space before the pipe) Fill in with heading / large-font text
Obsolete <title> <title>2020 admissions</title> but the page says "2021 admissions" Use the visible 2021 text
Inaccurate <title> The title says "Giant teddy bears" but the page changes dynamically Switch to text that reflects the actual content
Micro-boilerplate All pages across a season are <title>My TV show</title> Add "Season N" to distinguish them
No clear main title Multiple headings at the same level Use the first <h1>
Writing system mismatch A Hindi page has an English title Use prominent text from the page
Duplicate site name The title contains the site name but the site name is already shown in the top right Remove the duplicate in the title

2.4 Changed the title but it didn't take effect?

→ Google may have identified one of the issues above and overridden your title.Check the issue list, confirm the title is better relative to the user query. Or go to Help Community to discuss.

Part 3 — Snippet

3.1 Snippet sources

Primarily automatically generated from page content. When Google considers the meta description more accurate, it uses the meta description.Different queries may return different snippets.

3.2 The 3 major methods to control snippets

Method Effect
<meta name="robots" content="nosnippet"> Do not show snippets at all (also cannot be used as input for AI Overviews/Mode)
<meta name="robots" content="max-snippet:N"> Maximum N characters;0 = nosnippet;-1 = Google decides
<div data-nosnippet>...</div> Exclude only specific paragraphs

Conflict:nosnippet takes precedence over data-nosnippet.

3.3 Meta description 6 best practice

  1. Unique per page(use site-level for homepage, page-level for inner pages)
  2. Include page-specific key information: author / date / price / manufacturer — not just one sentence, can be structured tags
    <meta name="description" content="Written by A.N. Author, Illustrated by V. Gogh, Price: $17.99, Length: 784 pages">
  3. Large sites can generate them programmatically(news / product aggregators) — humans cannot write them all, programmatic generation is acceptable,quality is a prerequisite
  4. Use quality descriptions(comparison example):
    • ❌ Keyword list:"Sewing supplies, yarn, colored pencils, sewing machines"
    • ✅ Explanation + details:"Get everything you need to sew your next garment. Open Mon-Fri 8-5pm, located in the Fashion District."
  5. Do not use the same description for every news article
  6. summarize whole page Do not just quote one paragraph

A deep link = the "Read more" link in the snippet that links to a specific paragraph on the page:

  1. Content must be immediately visible(not hidden behind folds/tabs)
  2. Do not use JS to control scroll position
  3. History API / hash fragment When making changes do not remove the hash(deep linking relies on the hash)

Part 4 — Favicon

4.1 Implementation

<link rel="icon" href="/path/to/favicon.ico">

All 3 rel values work:

  • icon(HTML standard, recommended)
  • shortcut icon(historical compatibility)
  • apple-touch-icon / apple-touch-icon-precomposed(iOS-friendly)

href can be absolute or relative URLs, and can be cross-domain(CDN).

4.2 The 6 major rules

  1. One favicon per hostname — example.com and news.example.com can be different, but example.com/news cannot have an independent favicon
  2. Must be crawlable by Googlebot-Image + homepage must be crawlable by Googlebot
  3. Visually represents the brand
  4. Must be square + at least 8x8px(Recommended > 48x48px for good display across devices)
  5. Stable URL, do not change it frequently
  6. Must not contain inappropriate content(pornography / hate / swastika) → Google will replace it with the default icon

4.3 home page type

  • ✅ https://example.com(domain-level)
  • ✅ https://news.example.com(subdomain-level)
  • ❌ https://example.com/news(subdirectory-level)— Not supported

Part 5 — Sitelinks

5.1 What it is

Automatically generated multiple links from the same domain, clustered under one text result as shortcuts.

5.2 4 best practices (affect generation quality)

  1. Title + heading informative / relevant / compact
  2. Logical site structure + important pages linked from other relevant pages
  3. Internal link anchor text concise + relevant
  4. Avoid content duplication
  • Delete the page
  • or add noindex

Part 6 — Featured Snippets

6.1 What it is

A special reversed snippet (description appears before the title). Also appears in People Also Ask.

6.2 Cannot be actively marked up

Google Algorithms decide which page to elevate as a featured snippet.There is no way to manually apply.

6.3 The 2 opt-out paths

A. Block all snippets(including featured):

<meta name="robots" content="nosnippet">
  • data-nosnippet exclude specific segments (nosnippet takes precedence).

B. Block only featured snippets and keep regular snippets:

<meta name="robots" content="max-snippet:[小数字]">

Continuously decreasing until it no longer appears as featured.Not guaranteed, but the shorter it is, the less likely it becomes.

⚠️ Google does not provide a minimum length threshold for featured snippets (varies by language/platform).

→ Jumps directly to the corresponding paragraph (if the browser supports deep linking).When not supported jumps to the top of the page.

Part 7 — Enriched Search

7.1 What it is vs Rich Result

Rich Result Enriched Search
Trigger structured data structured data
Experience Enhanced snippet interactive / immersive(independent search interface)
Example review stars jobs search / recipe filter
  • Job Posting
  • Recipe
  • Event

7.3 5 major quality guidelines (all must be satisfied)

Guideline Meaning
Required properties Missing required attributes → not eligible
Completeness Providing more recommended properties → ranking improvement signal
Relevance Data must be truly relevant (a woodworking site marking recipes ❌)
Leaf content only Only detail pages are indexed,not listing pages(category pages)
Content policies Each type has its own content policies

⚠️ If a large proportion of the site fails to meet standards → the entire site is removed from enriched search.

7.4 Duplicate SD is not counted as duplicate content

Example: The same job listing displayed in multiple locations, with the same description but different locations → the enriched search algorithm does not treat it as duplicate.

Part 8 — AI Features (AI Overviews + AI Mode) ⭐

8.1 Key understanding

AI Overviews / AI Mode have no special SEO requirements. SEO fundamentals apply directly.There is no need to:

  • create machine-readable files
  • create AI text files (llms.txt etc.)
  • add special schema.org markup

See Google SEO Starter Guide + How Search Works + Search Essentials AEO/GEO/chunking/llms.txt are all myths.

8.2 AI Overviews vs AI Mode

AI Overviews AI Mode
Appears when automatically determined to provide additional value (often not triggered) when the user specifically queries in AI mode
Use case Quick gist of complex questions + jump links for exploration Exploration / reasoning / complex comparison
Link diversity More diverse than regular search Same
Technology query fan-out (search across multiple subtopics) Same

8.3 Eligibility to appear

The page satisfies:

  1. Indexable(can be shown in Search)
  2. Can display snippets(no nosnippet)
  3. satisfies Search technical requirements

That is all.No additional technical requirements.

8.4 SEO best practices (same as regular SEO)

  • robots.txt allows crawling
  • Discoverable via internal links
  • Good page experience
  • Important content is readable in textual form
  • High-quality images / videos
  • structured data Consistent with visible text
  • Merchant Center / Business Profile is up to date

8.5 Monitoring

Search Console Performance report → The "Web" search type includes AI Overviews + AI Mode traffic.

⭐ Clicks from AI Overviews are higher quality(users stay on the site longer).

8.6 Controlling content not to appear in AI

Control Effect
nosnippet / data-nosnippet / max-snippet Restricting snippets → also limits AI from using them as input
noindex If the entire page is not indexed, it will not appear in AI
Google-Extended robots.txt control Blocks use for Google other AI systems training / grounding (does not affect Search)

8.7 Troubleshooting preview controls not taking effect

  1. Use URL Inspection to check the HTML received by Googlebot
  2. Wait for recrawl after changes (days to months) — use URL Inspection to request acceleration

Part 9 — Translated Results

9.1 What it is

Google automatically translates the title link + snippet translates into the user's language for display.

9.2 21 supported languages

Arabic / Bengali / English / French / German / Gujarati / Hindi / Indonesian / Kannada / Korean / Malayalam / Marathi / Persian / Portuguese / Spanish / Tamil / Telugu / Thai / Turkish / Urdu / Vietnamese

Available on both mobile and desktop.

9.3 User click behavior

Jumps to the machine-translated page after Google Translate (there is an option to view the original).Google does not host translated pages — equivalent to accessing via the Google Translate entry / Chrome built-in translation. JS / images still work.

9.4 Opt out

<meta name="robots" content="notranslate">
<meta name="googlebot" content="notranslate">

or HTTP header X-Robots-Tag: notranslate.

9.5 Monitoring

Search Console Performance report Search Appearance filter → View translated impressions/clicks.

Part 10 — Ad Network Adaptation for Translated Pages (Technical)

10.1 Problem

User clicks a translated result → the page URL is rewritten by Google Translate (-.translate.goog)→ the URL received by the ad network is not the publisher's original URL → ad anomalies.

10.2 9-step decoding algorithm

proxyUrl: https://1-en--us-example-com.translate.goog/page?_x_tr_enc=1

1. 取 ".translate.goog" 前的 domain prefix
2. _x_tr_enc 按 "," split → encoding_list
3. 若 _x_tr_hp 存在 → 前置到 domain prefix
4. encoding_list 含 "1" 且 prefix 以 "1-" 开头 → 去掉 "1-"
5. encoding_list 含 "0" 且 prefix 以 "0-" 开头 → 去掉 "0-",设 is_idn=true
6. /\b-\b/ 替换为 "."
7. "--" 替换为 "-"
8. is_idn=true → 加 "xn--" punycode 前缀
9. (可选)转 Unicode

10.3 Complete JS implementation

function decodeHostname(proxyUrl) {
  const parsedProxyUrl = new URL(proxyUrl);
  const fullHost = parsedProxyUrl.hostname;
  let domainPrefix = fullHost.substring(0, fullHost.indexOf('.'));

  const encodingList = parsedProxyUrl.searchParams.has('_x_tr_enc')
    ? parsedProxyUrl.searchParams.get('_x_tr_enc').split(',') : [];

  if (parsedProxyUrl.searchParams.has('_x_tr_hp')) {
    domainPrefix = parsedProxyUrl.searchParams.get('_x_tr_hp') + domainPrefix;
  }

  if (encodingList.includes('1') && domainPrefix.startsWith('1-')) {
    domainPrefix = domainPrefix.substring(2);
  }

  let isIdn = false;
  if (encodingList.includes('0') && domainPrefix.startsWith('0-')) {
    isIdn = true;
    domainPrefix = domainPrefix.substring(2);
  }

  let decodedSegment = domainPrefix.replaceAll(/\b-\b/g, '.').replaceAll('--', '-');

  if (isIdn) {
    decodedSegment = 'xn--' + decodedSegment;
  }
  return decodedSegment;
}

10.4 Rebuild the complete URL

  1. Original path / fragment / params are preserved
  2. hostname is replaced with the decoded one
  3. Remove all _x_tr_* parameters

Part 11 — Preferred Sources (Top Stories Preference)

11.1 What it is

When a user selects a site as a "preferred source" in Source Preferences tool → that site is displayed more prominently for relevant news queries in Top Stories.

11.2 Applicability

  • Available globally
  • All Google Search languages
  • Only domain-level and subdomain-level sites are eligible — subdirectory (example.com/blog)is not allowed

11.3 How to drive users to select you

A. Push deeplinks to social media / websites

Format:

https://google.com/preferences/source?q=Your_Website's_URL

Example:https://google.com/preferences/source?q=example.com

B. Add the Google-provided button on your site

Button assets ZIP — 17 language versions, including Danish / English / French / German / Hebrew / Hindi / Japanese / Korean / Portuguese (Brazilian) / Russian / Spanish / Swedish / Turkish / Ukrainian, etc.

⚠️ This is optimization guidance, not a technical requirement — Google automatically determines whether your site is in the Source Preferences candidate pool.

Part 12 — Related wikis

Last updated

2026-05-17 — Compiled based on 11 articles in Google Search Central's appearance series (latest from 2025-12 to 2026-04). Fully includes AI Overviews / AI Mode 2026-04 documentation + Preferred Sources 2026-04 + Snippet 2026-04.

Sources and related materials