AI Afterhours

Most retailers have a forecasting problem they have been solving the same way since 2010. The data has changed. The signals have multiplied. The planning systems have not caught up. Somewhere underneath the dashboards and the spreadsheets, a 2026 business is still running on 2010 logic. Season 1 of AI Afterhours, Signals & Noise, is about how organizations cut through noise, find the signals that matter, and turn them into better decisions. In Episode 3, Demand Forecasting in Retail, I sat down with Bob and Varun for the final episode in our three-part retail arc. Harsha was out sick this week. Get well soon, Harsha. Bob opened with the frame that carried the whole conversation. The signal-to-noise ratio has flipped. Historical data used to get planners most of the way there. Now it is barely a starting point. He walked us through shifting planners from curation to exception management, and turning the planning team from a cost center into a value driver. Varun took it higher. When the underlying systems do not talk, the planning room becomes a war room where every team is defending a different number. When the data underneath is broken, the AI on top is just polished chaos. Bob's parting shot is the one to take with you. Stop trying to automate chaos. Fix the chaos first. Next week we pivot to the factory floor with Episode 4, Manufacturing with Dovient. Different pressures. Same core question. Bob, thank you. The bar you set is going to be hard to top. Thank you for tuning in! See you next episode.

In this episode of AI Afterhours: Signals & Noise, Sean Fleming sits down with Bob Ntuaremba Leyo, Harsha Varun, and Varun Vemula to break down what margin protection at scale actually looks like in retail.This conversation goes beyond markdowns and promotions to get at the real issue. Margin erosion starts much earlier with poor allocation decisions, centralized planning assumptions, disconnected execution, and delayed action across merchandising, pricing, and supply chain. Bob grounds the conversation in the business reality, showing how retailers lose margin through inventory misplacement, costly transfer corrections, and spray and pray promotions that discount purchases customers would have made at full price.Harsha explains why retailers need to move beyond point forecasts and start thinking in terms of probabilistic forecasting, simulations, causal modeling, and multi-echelon planning. He also unpacks why separating baseline demand from promotions, seasonality, weather, and external events is critical if teams want to make better planning decisions with confidence.Varun focuses on the gap between seeing a signal and acting on it. The discussion explores why legacy systems and siloed teams slow retailers down, how disconnected decisions compound margin loss, and why the companies that win are the ones making faster, more connected decisions in real time.The takeaway is simple: competitors are not just winning on price or promotion. They are winning on speed, visibility, and the ability to act earlier with trusted data.Be sure to tune in next week as we continue the retail series with a deeper look at demand forecasting. We will break down how better signals, stronger forecasting logic, and faster decisions help retailers plan with more confidence and protect performance at scale.

Retail planning is still broken in a very familiar way. Too much historical logic. Not enough live signal. Too much cluster thinking. Not enough operating truth.In Episode 1 of AI Afterhours: Signals & Noise, I sat down with Bob, Harsha, and Varun to break down why stockouts, overstocks, and missed margin are usually not isolated failures. They are the downstream consequence of weak signals, slow planning cycles, and planning systems that still assume stores behave the same when they do not. Bob made that real from the operator side. Harsha explained why forecast quality depends on better signals, simulation, and planner knowledge that usually never makes it into the system. Varun brought it back to the business. If you are still planning backward, you are probably bleeding margin forward.Next week we get more specific with Margin Protection at Scale. Same category. Same pressure. Closer to the money.Thank you for tuning in! See you next week.

A preview of Episode 1 of AI Afterhours featuring Bob Ntuaremba Leyo on why stores are not interchangeable. The same cluster on paper does not mean the same demand in reality. This clip gets into why local signals, hyper-localization, and operating truth matter if you want better planning, fewer stockouts, and less margin bleed.