Multi-Turn Conversation
Definition
Multi-turn conversation is the norm in sophisticated chatbot interactions. Unlike single-turn exchanges where each message is handled independently, multi-turn conversations require the system to track state across multiple rounds of dialogue. This enables the bot to handle tasks that require gathering multiple pieces of information, clarifying ambiguities, conducting follow-up questions, and maintaining a coherent narrative thread. Modern LLMs excel at multi-turn conversation because the full conversation history is passed as context with each API call.
Why It Matters
Most real customer service scenarios require more than one exchange. 'I need to change my delivery address' requires collecting the new address, confirming the order, checking if the order is still modifiable, updating the system, and confirming success — multiple turns. Without multi-turn capability, chatbots can only handle trivial queries, severely limiting their value in support operations.
How It Works
In LLM-based systems, multi-turn conversation is handled by passing the full conversation history (as a list of user/assistant message pairs) with each new API request. The model uses this history to understand references, maintain consistency, and build on prior responses. In rule-based systems, a conversation state object is passed between dialogue turns, updated with each collected value.
Real-World Example
User: 'I want to update my billing address.' Bot: 'Sure! What's the new address?' User: '123 Main St, Boston, MA.' Bot: 'Got it. Should I update this for all future invoices or just the next one?' User: 'All future ones.' Bot: 'Done! Your billing address has been updated.' — four turns, one coherent task.
Common Mistakes
- ✕Truncating conversation history too aggressively to reduce token costs, losing context that the model needs for accurate responses.
- ✕Not resetting state after task completion, causing bleed-over between different topics within the same session.
- ✕Building single-turn flows and calling them multi-turn — true multi-turn requires genuine context tracking, not just longer prompts.
Related Terms
Single-Turn Conversation
A single-turn conversation is a chatbot interaction where the user's message is fully resolved in one bot response, with no follow-up needed. Common for simple FAQ queries, single-step lookups, or informational requests where the complete answer can be provided immediately without requiring additional input from the user.
Dialogue Management
Dialogue management is the component of a conversational AI system that tracks conversation state and decides what the bot should do next — ask a follow-up question, retrieve information, take an action, or hand off to a human. It is the 'brain' that orchestrates a coherent, goal-directed conversation across multiple turns.
Contextual Awareness
Contextual awareness is a chatbot's ability to understand and remember information from earlier in a conversation — or from external sources like user profiles and page data — to give relevant, personalized responses. A context-aware bot doesn't treat each message as isolated but understands it as part of an ongoing interaction.
Chatbot Memory
Chatbot memory is the ability of a chatbot to retain and recall information across conversations — not just within a single session, but across multiple sessions over time. A chatbot with memory can greet returning users by name, remember their preferences, and pick up where previous conversations left off.
Slot Filling
Slot filling is the dialogue management process of collecting all the required pieces of information (slots) needed to complete a task. The chatbot systematically asks for any missing slots — like date, time, or account number — until it has everything needed to fulfill the user's request.
Ready to build your AI chatbot?
Put these concepts into practice with 99helpers — no code required.
Start free trial →