Blog.

Not All Products Behave the Same, So Why Forecast Them That Way? Imagine trying to forecast daily demand for toothpaste the same way you would for holiday gift baskets. One sells steadily year-round, the other spikes unpredictably. Yet in many businesses, all products are still forecasted using the same models, with the same assumptions, and often with the same level of effort. The result? Overstocked warehouses, stockouts during peak demand, and wasted hours spent fine-tunin

For decades, demand forecasting followed a relatively clear path. Forecasters relied on what customers expressed through purchases, search queries, and seasonal trends. These signals were observable, repeatable, and mostly predictable. Forecasting models based on historical data could produce reliable estimates. Supply chains could plan with a reasonable degree of confidence. That world is fading. Today, social media is reshaping how people discover products, interact with co

More and more businesses are using integrated AI and machine learning (ML) models within their ERP systems to forecast product demand but many struggle to explain these forecasts to management and colleagues. Why? Because AI and ML models are complex and operate in high-dimensional spaces, which are harder to understand than traditional models based on simpler, linear relationships. Even though these models have demonstrated that they can improve forecast accuracy, they are d

Intermittent demand where demand occurs sporadically with many periods of zero sales, presents a unique challenge for forecasters and supply chain planners. Traditional metrics for forecast accuracy can be misleading when applied to slow-moving items, and focusing solely on these can lead to suboptimal inventory decisions. This article explores which metrics to use for evaluating forecasts of intermittent demand, including accuracy measures (and their limitations) and invento

You forecast demand for 1,000 SKUs, and half are slow movers with lots of zeros. Your MAPE looks great, but your warehouse is still full, cash is tied up in inventory, and planners are losing trust in the numbers. Sound familiar? Intermittent demand is one of the most persistent and costly challenges in supply chain forecasting. It shows up in spare parts, seasonal or niche SKUs, luxury and slow-moving goods, and increasingly, in long-tail assortments in retail and e-commerce

The Game-Changer You’re Not Using... Yet Imagine identifying early signs of customer dissatisfaction before they decide to leave, how investors feel before they act, or what your employees truly think before they leave. Sentiment analysis is the key to unlocking these insights, yet many businesses still underutilize its potential. Sentiment analysis, also known as opinion mining, uses AI and machine learning to analyze text data and determine whether the sentiment is positive

Can social media predict product demand?  Many supply chain professionals still rely on traditional forecasting methods, but what if social media could give us real-time demand signals before sales even happen? In my recent research, we found that analyzing social media sentiment and engagement can improve demand forecasting accuracy by 42%, especially for new products with no historical data. Let’s break it down. Image generated by AI (DALL·E) The Challenge: Forecasting New