Forecasting demand for new products is very challenging.
Without historical data to draw upon, how do you predict the best way to allocate new products across a variety of channels?
Guess wrong and you will either:
- run out of inventory,
- stock too much inventory that will need to be marked down, or
- stock the right inventory in the wrong place (that will need to be transferred, or worse, marked down as well).
These are all very costly mistakes. AI can help to deal with the pitfalls of the traditional forecasting methods used by manufacturers or distributors, of which none are very accurate for predicting the performance of new products because they are only based on the past sales of similar products and not taking into account what is happening exactly with the sales of the new product.
There’s a way with AI, and it’s easier than you may think.
What will I learn?
You will learn from other company examples like Jacobs Douwe Egberts how:
- You can blend your previous market insights with the earliest possible market feedback for the new product
- Distinguish initial pipe fill from real sales
- Phase in and phase out can be done easily
- Understand cannibalization effects if they occur
- Understand the requirements of launch inventory
- To match the demand pattern as early as possible with known demand patterns

Whom will I listen to?
Anupam is 1 of 5 people that have gone through the onboarding data sets of almost all multi-national companies in the world and has a unique understanding of how new product introductions affect the demand planning processes.