Search Features & Rich Results Vintage

Bill Slawski analyzed patents that explain how Google creates and displays rich search features — the elements beyond the basic "ten blue links" that make modern search results visually rich and informationally dense.

Search Feature Architecture

Featured Snippets

Bill's 2022 analysis of a featured snippet patent revealed how Google scores and selects the text that appears in the featured snippet position.

Featured Snippet Answer Scoring

Scoring Components

Component	What It Measures
Query-dependent score	How relevant the answer passage is to the specific query terms
Query-independent score	Overall quality of the page and passage (authority, structure, clarity)
Answer completeness	Whether the passage fully answers the question
Passage clarity	How clearly the answer is stated
Source authority	Trustworthiness of the page providing the answer

How to Optimize for Featured Snippets (from Patent Insights)

1. Provide clear, direct answers within your content
2. Structure answers with clear headings followed by concise explanations
3. Use lists and tables when appropriate — these formats are extractable
4. Answer the question early in the relevant section
5. Maintain page quality — query-independent signals still matter

Source: Featured Snippet Answer Scores (2022)

Knowledge Panels

Knowledge Panels are triggered when Google recognizes a known entity in a query.

Knowledge Panel Trigger Process

What Makes a Knowledge Panel Appear

Based on patent analysis:

• The entity must exist in the Knowledge Graph
• Sufficient attributes must be available (name, description, image, type)
• The query must have clear entity-seeking intent
• Confidence must be high that the right entity is identified (disambiguation resolved)

Alternative Titles and Snippets

Bill analyzed a 2012 patent on how Google selects alternative titles and snippets for search results.

Dynamic Title and Snippet Selection

Google does not always use the page's title tag or meta description. Patents describe a system that:

1. Classifies query type — Different queries may need different title/snippet emphasis
2. Classifies page elements — Identifies headings, descriptions, body text, structured data
3. Selects the best representation — Chooses the title and snippet that best match the query
4. May rewrite snippets — Pulls text from the page that is most relevant to the query

Source: How Classification of Page Elements May Influence Alternative Titles and Snippets (2012)

Universal/Blended Search

Bill documented the evolution from "ten blue links" to blended search results that include multiple content types.

Blended Search Timeline

Google progressively integrated different search verticals:

• 2001 — Image Search launched separately
• 2002 — News Search added
• 2002 — Product Search (Froogle)
• 2005 — Google Maps
• 2007 — Universal Search launched (blending all verticals into web results)

How Blended Results Are Determined

The blending algorithm considers:

• Query type — Image-worthy queries get image packs
• Result quality — Only high-quality vertical results are blended in
• User intent — The likely user goal determines which verticals appear
• SERP diversity — Ensuring the results page provides varied information types

Source: 10 Most Important SEO Patents: Part 9 - From Ten Blue Links to Blended and Universal Search (2012)

Sitelinks

Sitelinks are the additional links shown beneath a main search result for branded/navigational queries.

How Sitelinks Are Selected

Based on patent analysis:

• Page importance — Internal link structure signals which pages are most important
• User behavior — Which pages users most frequently navigate to on the site
• Query relevance — Which sub-pages are most relevant to the query
• Content freshness — Recently updated pages may be prioritized
• Page uniqueness — Pages with distinct content (not similar to other sitelinks)

Search Suggestions

Patents describe the systems behind Google's autocomplete and related search suggestions:

Suggestion Generation

Source	Contribution
Query logs	Most popular completions of the typed characters
Trending queries	Recently popular searches
User history	Past searches by this user
Entity associations	Queries related to recognized entities
Geographic context	Location-relevant suggestions
Temporal context	Time-appropriate suggestions

Key Takeaways

1. Featured snippets are scored — Both query-dependent relevance and query-independent quality determine featured snippet selection.
2. Knowledge Panels require entity recognition — Getting into the Knowledge Graph is the prerequisite for a Knowledge Panel.
3. Google rewrites your titles and snippets — Your title tag and meta description are suggestions, not guarantees.
4. Universal search means competing with all content types — Your content competes with images, videos, news, and maps, not just other web pages.
5. Sitelinks are algorithmically selected — Internal linking structure and user behavior determine which sitelinks appear.
6. Structure content for extraction — Clear headings, lists, tables, and direct answers make content easier for Google to feature.