Measure retrieval quality separately before tuning prompts — 73% of RAG failures are in the retrieve step

Free Academy·low signal

2026 analysis consistently finds that when RAG fails, the failure is in retrieval 73% of the time, not generation — yet teams keep tuning prompts. The actionable move is to instrument and score the retrieve step in isolation (recall@k on the top 50–100 candidates before rerank) so you know whether the right chunk was even fetched. Combining dense vector search with sparse BM25 before reranking is the production standard because BM25 catches exact terms, names, and acronyms that embeddings compress away.