# Four answer engines walked into a bar. Cited four different bartenders. > Why a win in one answer engine tells me very little about the other three. Reddit is part of that story. By Redtail. The engines in Semrush’s sample drew on different source sets. I measure each separately and use Reddit alongside search, product documentation and other credible sources. ## How I approached it I compared the reported citation overlap in Semrush’s 5,000-keyword study. I kept domain-level and URL-level overlap separate, then mapped the findings to a repeatable brand-question audit. ## They are not in a group chat. Someone sends me one flattering AI answer and says: we’ve cracked it. I want the screenshot too. Then I want to ask the other engines. Semrush’s July 2025 study used 5,000 keywords and more than 150,000 unique citations. It balanced four search-intent groups and two search-volume bands, using desktop results. Against Google’s organic top ten, Perplexity had over 91% domain overlap and 82% URL overlap. AI Overviews had about 86% and 67%. AI Mode had about 54% and 35%. Those figures describe overlap with Google, not the percentage of times a brand appeared. Reddit was prominent across the AI surfaces studied. That supports investigating it in the channel mix. It doesn’t mean each engine reads the same thread or recommends the same product. Sources: [Semrush — AI Mode comparison study](https://www.semrush.com/blog/ai-mode-comparison-study/) ## A denominator can ruin a perfectly good hot take. Domain overlap is a broad signal: two systems can cite the same website while choosing completely different pages. URL overlap asks a narrower question. Neither number is the same thing as brand sentiment, recommendation quality or traffic. Google itself says AI Mode and AI Overviews may use different models and techniques, producing different answers and links. It also describes query fan-out, where related searches help assemble a response. A single neat keyword ranking cannot tell the whole story. My conclusion is operational: I need a question set and separate records for each engine. The bar chart can wait until the underlying rows make sense. Sources: [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features) ## Reddit gets a seat. So does the rest of the internet. I use Reddit to learn how people describe a problem before I turn it into a prompt. Their language tends to be less polite than a keyword tool. That is often helpful. For every important buying question, I look at the actual cited pages. A product-documentation answer needs a documentation fix. A misleading comparison needs better evidence. A community thread needs someone useful in the conversation. Those are different jobs. I connect them through one consistent product story: who it helps, what it does, what it costs and where it falls short. The website holds the detail. Relevant videos, coverage and community answers give people other ways to encounter it. ## My scorecard has rows. I know. Filthy behaviour. I would track the same set of problem, comparison and recommendation questions on a regular schedule. I’d record the full answer, whether the brand was named, whether it was recommended, the sentiment and every source URL. I’d also note whether the engine searched at all. A response without live sources should not get silently mixed into a citation-rate chart. Repeated runs matter because answers can vary. Then I would prioritise fixes by the questions that matter commercially and the evidence actually missing. Reddit could be the next move. So could a clear pricing page. The work follows the gap. - Keep each engine’s observations separate. - Distinguish a brand mention from a recommendation and a citation. - Record source URLs, prompt wording and dates. - Check enquiries and conversions outside the citation scorecard. ## Sources - [Semrush — AI Mode comparison study](https://www.semrush.com/blog/ai-mode-comparison-study/): Published July 21, 2025. 5K keywords; 150K+ unique citations. Observational snapshot. - [Google Search Central — AI features and your website](https://developers.google.com/search/docs/appearance/ai-features): Official explanation of query fan-out and differences between AI Mode and AI Overviews.