07.03 · Concept
The Off-Site Graph: Reddit, YouTube and Wikipedia
Name the third-party surfaces that answer engines cite most, decide which ones your product should legitimately be present on, and read the conflicting citation studies critically.
Answer engines cite more than your own site: they draw on off-site evidence from Reddit, YouTube, Wikipedia, publishers, forums, reviews, GitHub, documentation, directories and other trusted surfaces. The useful goal is legitimate corroboration where real users already compare, question, demonstrate or define your product.
What this lesson answers
- which sites do answer engines cite most
- should my product be on Reddit YouTube Wikipedia
- how to read AI search citation studies
Notes
Answer engines do not only read your website. They build answers from a wider off-site graph: search results, authoritative domains, user-generated discussions, videos, knowledge bases, product listings, reviews, and structured reference pages. In practice, surfaces like Reddit, YouTube, Wikipedia, major publishers, forums, review sites, GitHub, documentation hubs, and industry directories can matter because they are places where products, entities, and claims are independently described, compared, questioned, and cited.
The mental model is not “get backlinks” but “be legibly present where the…
Common questions
- Why do Reddit, YouTube and Wikipedia matter for AI search?
- They often contain independent descriptions, comparisons, demonstrations and disputes rather than only brand-written claims. Retrieval systems and answer engines can use those surfaces as corroborating evidence. Their value depends on whether the presence is genuine, useful and relevant to the query, not merely whether a brand has created an account or page there.
- Should every product try to appear on every cited platform?
- No. The right surfaces depend on where a real user would expect evidence for that product. A developer tool may need GitHub, documentation, technical blogs or Stack Overflow. A consumer product may need reviews, YouTube and Reddit. Forced promotion, fake reviews and low-quality community posts can reduce trust or be removed.
- Why do AI search citation studies disagree with each other?
- They often measure different things. One report may inspect Google AI Overviews, another may look at Perplexity, and another may infer mentions from ChatGPT-style responses. Query sets, regions, verticals, timing and the definition of a citation all change the results, so the method matters more than the headline ranking.
Short definition: what is Off-Site Graph: Reddit, YouTube and Wikipedia?