Google holds a patent literally titled "Contextual estimation of link information gain." Beginners can use the idea without the patent.
Evidence:
— The patent describes scoring documents by how much new information they add versus documents the user already saw — rewarding novelty, not repetition.
— SERP teardowns of mature informational queries show top results increasingly differentiate: one adds original data, another adds a tool, another adds first-hand testing. The pure rehash sits lower.
— Studies of AI Overviews and featured snippets show passages with unique specifics (numbers, named methods, first-hand observations) get pulled more often than generic summaries.
Nuanced conclusion: information gain is the practical antidote to "write what's already ranking." Before publishing, ask: what does my page contain that the top 10 don't? An original screenshot, a small dataset, a tested result, a counter-example. If the answer is "nothing," you're adding noise.
Method note: based on the granted Google patent (US10970293) and snippet/AI-Overview citation observations; patents indicate intent, not confirmed live weighting.
Confidence: medium
The Authority Files
@AuthorityFiles
What is "information gain" and how do I get it?
This post was published in the Telegram channel The Authority Files (@AuthorityFiles). All its posts are on the channel page.
Open on Telegram