Brass SEO logo
Brass-SEO
BY COPPER SUN

Ranking Signals & Google's Algorithms

Brass-SEO · 4 entries · last verified August 2026

Brass-SEO cites these primary sources when explaining ranking signals to users. Google's own documentation, the original PageRank paper, and confirmed first-party data from the 2024 documentation incident are the only sources here — no inference, no competitor analysis.

Contents — 4 entries
  1. 1.A Guide to Google Search Ranking Systems
  2. 2.The PageRank Citation Ranking: Bringing Order to the Web
  3. 3.2024 Google Search Documentation Leak — Content Warehouse API
  4. 4.How Google Search Works
  5. —Frequently Asked Questions

A Guide to Google Search Ranking Systems

Google Search Central. Maintained by Google.

Brass-SEO draws on this as the primary public catalog of Google's active ranking systems. The guide names 17 ranking systems currently in use — including BERT (natural language query understanding), Helpful Content System (site-level quality assessment), Neural Matching (concept-level query matching beyond keywords), Passage Ranking (relevance within long documents), and Freshness (time-sensitive content prioritization). This is a first-party Google disclosure: these systems are confirmed active, not inferred from testing. Brass-SEO uses this list to explain why ranking difficulty varies by query type — a news article competes primarily on freshness criteria, while a service page competes primarily on Helpful Content System criteria.

Examines:
Google's publicly maintained list of 17 named ranking systems — what each does, the query types it primarily affects, and its confirmed active status.
Brass-SEO draws on:
The 17 named ranking systems — the basis for Brass-SEO's query-class-specific recommendations rather than generic ranking advice that treats all queries identically.

The PageRank Citation Ranking: Bringing Order to the Web

Brin & Page, 1998/1999. Stanford InfoLab Technical Report.

Brass-SEO treats this as the foundational technical document for understanding link-based authority. Brin and Page's Stanford technical report describes PageRank — the algorithm that treats hyperlinks as votes and weights them by the authority of the linking page, computing a recursive score across the web graph using a damping factor of 0.85 as the standard parameterization. While Google has built 17 or more named systems on top of this foundation, the core concept of link-derived authority has not been abandoned; it has been augmented. The paper remains the primary source for understanding why internal linking architecture, link equity flow, and backlink diversity still drive organic authority signals.

Examines:
The original PageRank algorithm — mathematical definition, the 0.85 damping factor as the standard parameterization, link-authority propagation, and the web-as-citation-network model.
Brass-SEO draws on:
The link-authority propagation model and damping factor — the conceptual foundation cited in Brass-SEO's internal linking and authority recommendations.

2024 Google Search Documentation Leak — Content Warehouse API

Google (confirmed authentic). Analysis by Erfan Azimi, Mike King (iPullRank), and others. May 2024.

Brass-SEO monitors this for documented signal evidence. In May 2024, a Google GitHub repository accidentally exposed 14,014 attributes of the Content Warehouse API — Google's internal document representation — and Google confirmed the data was authentic. Analysis by independent SEO researchers identified NavBoost click signals (goodClicks, badClicks, and lastLongestClicks tracked over a 13-month window), a siteAuthority score as an explicit site-level attribute, Chrome clickstream data as a quality input, and hostAge as a factor that appears to affect content sandboxing for new domains. These are attribute names documented in Google's confirmed internal system, not signals inferred from external testing.

Examines:
The 14,014-attribute Google Content Warehouse API documentation accidentally exposed in May 2024 and confirmed authentic by Google — analyzed by independent SEO researchers.
Brass-SEO draws on:
The NavBoost click signals, siteAuthority attribute, and hostAge factor — cited when Brass-SEO discusses click-through rate quality, site authority, and new-site ranking timelines.

How Google Search Works

Google Search Central. Maintained by Google.

Brass-SEO cites this when explaining Google's three-stage process to new users. Google's documentation describes search in three confirmed stages: crawling (discovering URLs via links, sitemaps, and redirects), indexing (parsing page content into the search index with associated signals), and serving (ranking indexed pages for a query using hundreds of factors). Notably, Google explicitly states that a page can be crawled but not indexed — the 'Crawled, currently not indexed' status visible in Google Search Console — and that not all crawled pages are served to users. This three-stage model is the basis for Brass-SEO's GSC status interpretation.

Examines:
Google's official explanation of the three confirmed stages of search — crawling, indexing, and serving — with descriptions of each stage and what can prevent a page from advancing.
Brass-SEO draws on:
The three-stage crawl → index → serve model — the basis for Brass-SEO's GSC status interpretation, helping users understand why crawling success does not guarantee indexing.

Frequently Asked Questions

How many named ranking systems does Google officially acknowledge?

Google's Guide to Ranking Systems lists 17 named systems that are confirmed active as of the most recent update. These include BERT, Helpful Content System, Neural Matching, Passage Ranking, RankBrain, Freshness, and others. The guide distinguishes these from updates (one-time changes) and core updates (broad periodic adjustments to ranking systems). The 17 listed systems represent Google's first-party disclosure — not a comprehensive list of all internal ranking signals.

What is NavBoost and what is its source?

NavBoost is a click signal system identified in the 2024 Google Content Warehouse API documentation leak that Google confirmed as authentic. The leaked attributes include goodClicks, badClicks, and lastLongestClicks — signals tracked over a 13-month window. These attribute names indicate that click behavior (including long-click satisfaction signals) is a documented input into Google's internal document representation. NavBoost was not previously publicly confirmed as a named system.

Does the original PageRank algorithm still matter in 2026?

Link-based authority signals remain active in Google's system. The Google Ranking Systems Guide lists 17 confirmed active systems, several of which build on link-derived authority. The 2024 Content Warehouse API leak documented a siteAuthority attribute as an explicit site-level score in Google's internal representation. The original PageRank model — treating hyperlinks as authority votes, weighted by the authority of the linking page using a 0.85 damping factor — remains the conceptual foundation, though it operates alongside 17 or more additional named systems.

What is the difference between crawling, indexing, and serving in Google Search?

According to Google's 'How Search Works' documentation, crawling is the process of discovering URLs through links, sitemaps, and redirects. Indexing is parsing page content and storing it in Google's search index with associated signals. Serving is ranking indexed pages for a specific query using hundreds of factors. These are separate stages — a page can be crawled but not indexed (shown as 'Crawled, currently not indexed' in Google Search Console), and an indexed page may still not rank well for a specific query.

Back to SEO & GEO Research Index