Introduction
TCS Optumera is TCS's AI-powered retail merchandising suite. It helps retailers forecast demand, plan assortments, set prices and markdowns, and allocate shelf space. In grocery, where margins are thin and thousands of SKUs change every week, the promise is large: better forecasts, fewer stock-outs and less waste. Yet a common problem with AI merchandising tools is that category managers override or ignore the recommendations. If a category manager for dairy rejects 60% of Optumera's price suggestions, the client sees little value and the renewal is at risk. The question asks how to improve Optumera so that grocery category managers actually accept its recommendations. I'll clarify, segment users, map the journey, analyse pain points, generate and prioritise solutions, and define metrics.
Adoption problems in AI products are usually trust problems. Structure your answer around the three reasons people reject a recommendation: they don't understand it, they don't believe it, or it conflicts with how they are measured.
Step 1
Clarifying Questions (5 mins)
- Which modules? Assume assortment (which SKUs to stock by store cluster) and pricing (regular price and promotion recommendations).
- Which client? A large grocery chain with 500+ stores, for example a US or UK supermarket, or a large Indian modern-trade retailer.
- What is the current acceptance rate? Assume about 35% of recommendations are accepted unchanged, 25% edited and 40% rejected.
- What is the goal? Increase acceptance of good recommendations, not blind acceptance. The real outcome is margin and availability improvement.
- Constraints: The client's data quality varies; we cannot rebuild the forecasting engine in this cycle.
Practice similar questions
Subscribe to access the full answer
.png){kind=logo}
