One of the most useful AI stories in marketing this week did not come from a campaign, a chatbot, or a creative tool. It came from pricing and promotion planning. A new arXiv paper published on June 16, 2026 describes two large-scale systems deployed at PepsiCo to support revenue growth management: PromoAI for promotion calendars and PricingAI for base-price optimization. That matters because it points to a more serious version of applied AI in marketing. Instead of asking how to generate more content, the systems ask how to make better commercial decisions across products, channels, timing windows, constraints, and margin goals.
For senior marketers, ecommerce leaders, and commercial directors, this is an important shift in emphasis. A lot of AI discussion in marketing still revolves around faster production: cheaper copy, more variants, more images, more automation. Those tools have value, but they often sit near the surface of the operating model. Pricing and promotion optimization sits much deeper. It influences what gets sold, when demand is stimulated, how margin is defended, and whether media pressure translates into profitable growth rather than noisy volume.
The AI opportunity is moving closer to revenue design
The PepsiCo paper outlines a system where machine-learning forecasts and optimization logic are used to search through huge numbers of price and promotion combinations. The interesting lesson is not the mathematics alone. It is the management implication. In many organizations, pricing, trade promotion, ecommerce, performance media, and brand planning still operate as loosely connected domains. That separation creates inefficiency. Marketing pushes demand. Sales teams negotiate conditions. Revenue-management teams protect margin. Analytics teams report after the fact. AI-driven optimization becomes valuable precisely because it forces these layers into a more unified decision structure.
That is why this topic matters beyond large enterprise CPG. The same principle applies to retailers, marketplaces, and ecommerce brands of many sizes. When acquisition costs are volatile and consumer demand remains sensitive to price, discounting, and timing, better creative alone cannot protect growth. Teams also need stronger decision systems for when to promote, how far to discount, which products to feature, and what trade-offs they are accepting between revenue, margin, and inventory movement.
Source:
arXiv: PepsiCo Deploys AI-Driven Pricing and Promotion Optimization at Scale
