A chatbot is software that simulates conversation. Older bots follow decision trees and keyword rules; modern ones use a large language model to interpret free-form questions and generate replies.
Why it matters
Chatbots handle common support questions at any hour, qualify leads, and guide users through tasks. They also create risk when they answer wrongly, leak data, or fail when the backend is down, so they need monitoring and a path to a human. Review real conversation logs regularly to find questions the bot handles badly.
Example
A rule-based bot matches keywords:
if "refund" in message: reply(REFUND_POLICY_TEXT)An LLM-based bot instead receives the conversation plus company documents and writes a tailored answer, often using RAG.
Common mistakes
- Offering no way to reach a person when the bot cannot help.
- Letting the bot make promises, such as refunds or prices, it is not authorized to make.
- Not monitoring the chat endpoint's availability and latency.
Related
Get notified when this term updates
Plus occasional practical tips on uptime and monitoring, nothing else. Unsubscribe anytime.
This page is general information written by our editorial team and updated when documentation changes. Servers and providers differ, so test changes safely first. See our editorial policy and disclaimer.