From Simulation to Strategy: Automating Personalized Interaction Planning for Conversational Agents

TL;DR

This work investigates how user personas shape the effectiveness of sales-oriented dialogue agents and demonstrates that occupation information most strongly influences conversational outcomes. A lightweight, occupation-conditioned strategy is derived from simulated user behaviors and integrated into an LLM-based SalesAgent using chain-of-thought prompting, without model fine-tuning. The study systematically analyzes age, gender, and occupation effects via large-scale user simulations, showing improved success rates and shorter dialogues when occupation-based targeting is applied, with robust generalization to unseen simulators. These findings support scalable, persona-informed dialogue customization for persuasive, business-oriented agents.

Abstract

Amid the rapid rise of agentic dialogue models, realistic user-simulator studies are essential for tuning effective conversation strategies. This work investigates a sales-oriented agent that adapts its dialogue based on user profiles spanning age, gender, and occupation. While age and gender influence overall performance, occupation produces the most pronounced differences in conversational intent. Leveraging this insight, we introduce a lightweight, occupation-conditioned strategy that guides the agent to prioritize intents aligned with user preferences, resulting in shorter and more successful dialogues. Our findings highlight the importance of rich simulator profiles and demonstrate how simple persona-informed strategies can enhance the effectiveness of sales-oriented dialogue systems.

Figures (4)

• Figure 1: Simulation overview.
• Figure 2: Comparison of intent and success intent distributions across different user occupations in SalesAgent. Transparent bars represent overall intent distributions, while solid bars indicate success intent distributions.
• Figure 3: Pipeline of SalesAgent w/ or w/o the occupation-based strategies.
• Figure 4: SalesAgent's intent distributions and success intent distributions w/o vs. w/ strategies when interacting with unseen users in the testing phase. Transparent bars represent overall intent distributions, while solid bars indicate success intent distributions.