Secrets from the Algorithm: Google Search’s Internal Engineering Documentation Has Leaked - iPullRank

🌈 Abstract

The article discusses how Google spokespeople have allegedly misled and misdirected the SEO community about various aspects of how their search systems operate, in an effort to control how SEOs behave. It presents assertions from Google employees alongside facts from documentation to highlight discrepancies between their public statements and the actual systems and metrics used by Google.

🙋 Q&A

[01] Google's Stance on Domain Authority

1. What is Google's stance on using domain authority as a ranking factor?

• Google spokespeople have repeatedly stated that they do not use "domain authority", implying they do not measure the overall authority or importance of a website.
• However, the article presents evidence that Google does have an internal metric called "siteAuthority" that is part of their Compressed Quality Signals and is used in their ranking systems.
• The article suggests that Google's statements on this topic are intentionally misleading or obfuscating the truth about their use of domain-level authority signals.

2. How does the article characterize Google's public statements on this issue?

• The article describes Google's statements as "misdirection and obfuscation" that allow them to "never directly answer the question" of whether they use sitewide authority metrics.
• It suggests Google is engaging in "gaslighting" and "confusion-by-way-of-semantics" to mislead the SEO community.

[02] Google's Use of Click Data

1. What does the article say about Google's use of click data and user signals in their ranking algorithms?

• The article presents evidence from Google patents and documentation that Google does in fact use click-based signals, including metrics like "bad clicks", "good clicks", "last longest clicks", etc. in their "NavBoost" ranking system.
• This contradicts public statements from Google employees like Gary Ilyes, who have dismissed the importance of click-through rate and other user signals.
• The article argues that Google is intentionally misdirecting the SEO community about the role of click data in their algorithms.

2. How does the article characterize Google's public stance on this issue?

• The article states that Google's spokespeople have been "adamant that there is no sandbox" for websites based on age or trust signals.
• However, the article presents evidence from Google's own documentation that there is in fact a "sandbox" used to segregate "fresh spam" websites.
• The article suggests that Google is being deceptive in denying the existence of this sandbox mechanism.

[03] Google's Use of Chrome Data

1. What does the article say about Google's use of Chrome data in their ranking algorithms?

• The article presents evidence from leaked Google documentation that their systems do in fact incorporate data from Chrome, contradicting previous public statements from Google employees like Matt Cutts and John Mueller that they do not use Chrome data for organic search rankings.
• Specific modules in Google's systems are shown to include "site-level measure of views from Chrome" and other Chrome-related attributes.

2. How does the article characterize Google's public stance on this issue?

• The article states that Google's public statements denying the use of Chrome data for organic search are false, and that the leaked documentation clearly shows they do incorporate Chrome data into their ranking algorithms.

[04] Panda Algorithm and User Signals

1. What does the article reveal about how the Panda algorithm works?

• The article cites a Google patent that indicates Panda is based on a scoring modifier calculated from two main signals: the count of independent links pointing to a site/subdomain/subdirectory, and the count of "reference queries" (i.e. previous search queries that have been associated with that content).
• This suggests Panda updates are driven by changes in the link graph and in the set of queries driving traffic to a site, rather than being a complex machine learning system as previously thought.

2. How does this new understanding impact SEO strategies?

• The article states that to recover from Panda and similar updates, the focus should be on driving more successful clicks using a broader set of queries, and earning more diverse backlinks.
• This aligns with the signals the article says Panda is based on - increasing qualified traffic and link diversity.

[05] Google's Measurement of Expertise, Authoritativeness and Trustworthiness (E-E-A-T)

1. What does the article reveal about how Google measures E-E-A-T?

• The article states that Google's documentation shows they explicitly store information about the authors associated with web pages, and have mechanisms to determine if an entity on the page is the author.
• This, combined with Google's use of entity embeddings and other signals, suggests they have a more comprehensive way of measuring the expertise, authoritativeness and trustworthiness of content creators.

2. How does this impact SEO strategies around E-E-A-T?

• The article indicates that focusing on building a strong brand and driving qualified traffic from a broad set of queries are important for demonstrating E-E-A-T to Google's systems.
• It suggests that authorship signals, while previously seen as limited, may be more valuable than assumed based on this new information.

[06] Other Ranking Factors Revealed

1. What other ranking factors does the article discuss?

• The article mentions several other signals and metrics used by Google, including:
  • "sourceType" - a measure of how valuable a page is based on where it is indexed in Google's tiered storage system
  • Metrics to detect spikes in "spam anchor text"
  • Tracking of page version history and "Nearest Seed" PageRank
  • Measuring average weighted font size and anchor text length
  • Scores for "original content" and "keyword stuffing"
  • Signals around freshness and domain registration information

2. How does the article suggest SEOs should respond to these revelations?

• The article states that SEOs should reconsider their link building strategies based on the more nuanced understanding of how Google values links.
• It also emphasizes the importance of creating high-quality, relevant content and building a strong brand presence, rather than focusing on gaming specific signals.