Seller improves revenue by jointly setting price and ad with AI
Marrying Pricing and Advertising with LLMs
Computer Science and Game TheoryMachine Learning
Summary
The paper studies how a seller can better sell products by choosing both the price and an advertisement written by a large AI language model. The demand for the product depends on both the price and the ad, but the seller only sees if a product was bought or not each time. To solve this, the authors propose a method that learns over time to pick the best price and ad together, with the AI generating ads and a model estimating the chance of purchase to guide pricing. Their approach showed noticeable revenue improvements in several simulated scenarios and a marketplace data simulation compared to simpler methods.
What this means in practice
- •For e-commerce pricing teams: Optimize product prices and AI-generated advertisements together to increase sales revenue in online stores.$Commercial implications: Enables development of software products that dynamically generate ads and prices to boost e-commerce revenue.
- •For digital marketing teams: Improve advertisement effectiveness by jointly adjusting ad content and pricing based on customer purchase feedback.
Authors
Alessandro Barro, Francesco Bacchiocchi, Francesco Emanuele Stradi, Alberto Marchesi
Abstract
We study a sequential pricing problem in which a seller jointly posts a price and an advertisement generated by a large language model (LLM). The seller aims to maximize revenue under an unknown product demand that depends on both decisions, while observing only whether each offer leads to a purchase. We propose an online actor-critic algorithm that combines low-rank adaptation (LoRA) of a pretrained LLM with a demand model fitted to available data. At each round, the actor generates an advertisement, and the critic estimates purchase probabilities to guide price selection. Then, the resulting feedback is used to update both the actor and the critic, with the critic's revenue estimates providing a baseline for policy gradient updates of the actor. To evaluate our approach, we develop an evaluation framework with three synthetic demand models and a demand simulator built from real-world marketplace data. Finally, we compare our algorithm with benchmarks that do not jointly optimize price selection and advertisement generation, achieving expected revenue gains over the reference policy of 5.69%, 5.18% and 55.96% under the three synthetic demand models and 5.81% under the marketplace simulator.