# Hershey uses AI to position s'mores ingredients together in-store, lifting seasonal chocolate sales 15% *The company deploys machine learning to predict optimal shelf placement for its chocolates alongside marshmallows and graham crackers during peak season.* By **Jenny Huang Goodman MPA MSc MHSA, Principal** — The Stash Edge, Hako Shikin LLC. Published 2026-08-24. Canonical: https://www.pops4.com/stash/articles/hershey-2026-08-24t09-7 Subject: Hershey Tags: ai merchandising, shelf placement, seasonal strategy, retail partnerships, basket analysis, complementary products --- The Hershey Company deployed AI-powered merchandising tools to increase sales of s'mores ingredients during peak camping and grilling season, according to Modern Retail. The system analyzes store-level data to recommend placing Hershey's chocolate bars adjacent to marshmallows and graham crackers, treating the three products as a single purchase occasion rather than separate categories. The company reported a **15%** lift in chocolate sales at stores that implemented the AI-recommended adjacencies compared to control locations. The AI tool ingests multiple data streams: foot traffic patterns by hour and day, local weather forecasts, regional grilling season timing, and historical basket data showing which shoppers buy all three s'mores components versus single items. It then generates store-specific planograms that position the products within arm's reach of each other, often in seasonal endcaps or near outdoor living sections. Hershey's sales teams share these planograms with retail buyers as data-backed merchandising recommendations rather than simple product placement requests. The mechanism works because it collapses decision friction. A shopper planning a camping trip who sees chocolate bars isolated in the candy aisle must remember to navigate to two other sections for marshmallows and crackers — a sequence that fails roughly **40%** of the time, according to Hershey's basket analysis cited in the report. When all three items appear together, the purchase rate for the complete s'mores set increases, and the average transaction includes more units of each component. The AI layer adds predictive timing: it flags which stores should build the displays two weeks before local grilling season peaks, based on historical sales curves and real-time weather data. The broader insight is that AI merchandising pays for physical products when it solves a verified purchase barrier with store-level precision. Hershey is not inventing demand for s'mores. It is removing the friction that causes shoppers to abandon incomplete purchases, and doing so with geographic and temporal accuracy that static planograms cannot match. A small brand selling a physical product that pairs with something else — coffee and biscotti, hot sauce and chips, candles and matches — can run the same play at local retail scale. Start by identifying your natural product pair from actual basket data or direct customer feedback, not assumption. If you sell at farmers markets or small grocers, propose a joint display to the buyer: your product and the complementary item together, backed by a simple sell-through promise. Offer to build and maintain the display yourself for the first month. Track daily sales of both products and share the results with the buyer weekly. If the pair lifts sales of the retailer's existing product, you earn expanded placement. For online DTC, create a bundle at a **10-15%** discount that includes your product and the pair item, either sourced wholesale or via affiliate link. In product photography and email, show the two items in use together, not separately. In cart abandon sequences, trigger a message when someone buys your product without the pair: "Most customers also grab [pair item] — add it now for [discount]." Track bundle attach rate and pair purchase rate as your core metrics. If the pair item is carried by a retail chain you want to enter, bring the bundle conversion data to the buyer meeting as proof that your product drives incremental sales of their existing inventory. The play scales when you shift from asking a retailer to stock your product to showing them how your product increases sales of something they already carry. Hershey framed its AI tool as a way to sell more marshmallows and crackers, not just more chocolate. A small brand can do the same without AI by manually tracking which products move together and proposing the adjacency with proof. ## The takeaway AI or not, pair your product with a complementary item the retailer already stocks, then prove the adjacency lifts sales of both. --- ## Publisher **Hako Shikin LLC** — Virginia Beach, Virginia. Founded 1997. ASI 217876 · DUNS 18-204-6339. Principal and author: **Jenny Huang Goodman MPA MSc MHSA**. - Author: https://www.huanggoodman.com/about - LLM context: https://www.pops4.com/stash/llms.txt - MCP endpoint, for AI agents: https://mcp.pops4.com/mcp - Client dashboard: https://dashboard.pops4.com/ - Catalogue: 70,000+ products, 200+ brands