Home/ Glossary/Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG)
A technique where an AI agent retrieves passages from your own knowledge base before answering, so replies are grounded in your documented policies and product facts rather than invented.
One term from the WhatsApp Business Platform glossary. Last updated 2026-07-25.
A language model on its own answers from what it absorbed during training, which is the wrong source for a question about one business's refund window or delivery timelines. Retrieval-augmented generation inserts a lookup step: the agent searches the business's own documents first, then answers using what it found.
For a support agent on WhatsApp this is the difference between a plausible answer and a correct one. It also makes the system auditable — a grounded reply can be traced to the passage it came from, and an agent that retrieves nothing relevant can be made to hand the conversation to a human instead of inventing an answer, which is the failure mode that makes teams distrust AI in support.
Reference: https://en.wikipedia.org/wiki/Retrieval-augmented_generation
Related terms
- Customer service window
- The 24-hour window that opens when a customer messages your business. Replies you send inside it are free-form and are not charged as template messages.
- WhatsApp Business Platform (WhatsApp Business API)
- Meta's programmatic interface that lets medium and large businesses send and receive WhatsApp messages at scale. It has no app interface of its own — you access it through a platform such as BotPulsar, and your business owns the underlying WhatsApp Business Account.
- Quality rating
- The health score Meta assigns each WhatsApp phone number, derived from how recipients react to its messages. Blocks and reports push it down, and a low rating restricts the number's messaging limit.
See Retrieval-augmented generation (RAG) in a real workspace
BotPulsar runs the official WhatsApp Business Platform with unlimited team members on every plan, starting at ₹0.