Precision
Precision is a retrieval metric measuring what fraction of the passages a system returned were actually relevant, capturing how much of the result set is signal versus noise.
Precision asks: how much of what was retrieved is actually relevant? It's the share of returned passages that genuinely answer the query. High precision means the result set is clean and on-point; low precision means it's padded with loosely-related noise. It's the natural counterpart to recall.
Precision matters to AEO because answer engines work from a tiny, high-precision set — reranking exists to push precision up before the model writes, keeping only the few best passages. To be one of them, your content has to be unambiguously relevant to the question, not just adjacent to the topic. That's the extractability pillar again: a passage that precisely answers one question survives the precision filter, while a vaguely on-theme one is trimmed away.
Example. If an engine returns five passages and four truly answer the question, precision is 80%. Reranking aims to make those final few nearly all relevant — so only the sharpest, most on-target passages, ideally yours, remain.
Relevant pillar
Related terms
- RecallRecall is a retrieval metric measuring what fraction of all the relevant passages a system actually found, capturing whether the right content was retrieved at all.
- RerankingReranking is a second pass in retrieval where an initial set of candidate passages is reordered by a more precise relevance model, deciding which few actually make it into the AI's answer.
- RetrievalRetrieval is the step where an AI system searches an index to find the most relevant passages for a query before generating an answer, and it decides which content is even eligible to be cited.