SPAR: Experience-Driven Strategy Evolution in Multi-Session B2B Sales
Hangming Zhang, Tengyue Ma and colleagues at Alibaba Group and South China University of Technology present SPAR, a self-evolving sales agent that treats selling strategies as testable hypotheses and revises them without parameter updates. The paper is accepted to the EMNLP 2026 Industry Track.
Ask this paper
Double loop. An inner loop adapts to each merchant through episodic memory across sessions. An outer loop maintains a shared strategy playbook and updates it through causal attribution and hypothesis-driven evolution.
Attribution. Outcomes are not treated as equally informative. SPAR separates failures caused by a flawed strategy from failures caused by external factors before it changes the playbook.
Live result. In an online A/B test on Alibaba's 1688 platform with over 200,000 merchants, SPAR increased conversion by 15.8% and high-intent merchant growth by 35.8%.
Offline results. In simulation it beats the strongest baseline by 9.6% on objection recognition and 5.5% on evidence support, and domain experts rate it 4.62 out of 5 on overall sales quality, the highest of all methods.
Limits. The playbook needs several interactions before it improves, so early merchants do not benefit, and individual LLM-driven playbook updates can be wrong.
Abstract
B2B outbound sales involves high transaction values and relationship-based trust, so closing a deal typically requires multiple rounds of outreach building on prior context. Existing LLM-based sales agents operate within single sessions and cannot carry forward context or refine strategies from experience. Methods that do learn from experience store lessons as unstructured text and treat all outcomes as equally informative, with no mechanism to distinguish a flawed strategy from an external factor. We propose SPAR, a self-evolving framework that models selling strategies as testable hypotheses and refines them without parameter updates. A double-loop architecture lets an inner loop adapt to individual merchants through episodic memory while an outer loop evolves a shared strategy playbook through causal attribution and hypothesis-driven evolution. In a live A/B test on the 1688 platform with over 200,000 merchants, SPAR achieves +15.8% conversion lift and +35.8% high-intent merchant growth. Human evaluation by domain experts rates SPAR at 4.62 out of 5 on overall sales quality, the highest among all methods.