Pages that readers commonly view one after another (series, guides split across pages, hub then spoke) share large blocks of the same introductions, definitions and advice. Google's information gain patent describes scoring documents by what they add beyond documents the user has already viewed. Google has not confirmed use in Search, so this is a proxy for a patented method, and it is also a plain reader experience problem.
Why does pages in a reading sequence repeat what the previous page said (information gain) matter for AI search?
Google's information gain patent describes scoring documents by what they add beyond documents the user has already viewed. Google has not confirmed use in Search, so this is a proxy for a patented method, and it is also a plain reader experience problem.
How do you fix it?
Give each page in the sequence a distinct job. Keep shared definitions on one page and link to it, cut repeated introductions, and add something new to each later page: data, examples or next steps.
How do you find it on your site?
- From analytics, list the most common page-to-next-page paths for content pages.
- For each pair, extract the main content of both pages (excluding navigation and footer).
- Measure sentence-level or shingle overlap between the second page and the first.
- Flag pairs where a large share of the second page's content restates the first.
Which ranking and citation factors does it touch?
Written for the Information gain patent page (evidence: Patent only). Google has not documented this system in a way a site can test directly, so this check measures a proxy for it, not the system itself.
What is the impact?
Indirect. Estimate only: benefits are to reader engagement and any novelty-based selection, neither directly measurable.
What is the evidence?
US11354342B2, Contextual estimation of link information gain (Google Patents).
Frequently asked questions about pages in a reading sequence repeat what the previous page said (information gain)
What does "Pages in a reading sequence repeat what the previous page said (information gain)" mean?
Pages that readers commonly view one after another (series, guides split across pages, hub then spoke) share large blocks of the same introductions, definitions and advice.
How do you fix pages in a reading sequence repeat what the previous page said (information gain)?
Give each page in the sequence a distinct job. Keep shared definitions on one page and link to it, cut repeated introductions, and add something new to each later page: data, examples or next steps.