Every new product launch presents major challenges in terms of forecasting demand levels over time. The reality is that the lifecycle of a new product is the ultimate unknown from a demand planning perspective. But that’s no excuse for not developing a plan.
Planning for difficult to forecast demand
The short-term lifecycle of new products is difficult to plan and therefore vital to do so in order to avoid costly excess product or lost sales opportunities.
In our view a New Product Introduction (NPI) plan should account for both:
- Sales volume over the long term at normal operating capacity (otherwise known as the baseline demand)
- The demand profile during the initial launch period
As part of SO99+, we offer stochastic (probabilistic) forecasting to predict the baseline demand, using techniques such as advanced attribute clustering to find patterns of historic demand from similar products. In addition, machine learning can predict the demand profile during the initial launch period.
Our models can include early indicators such as web analytics, product attributes or even social media data, which enable us to predict the performance and profile early on of a product launch. These inputs are hard to account for with traditional forecasting techniques, but can be automatically incorporated using machine learning embedded in the demand planning software where we leverage deep neural networks to produce a forecast that is responsive to early market indicators.
When the bulk of sales revenue occurs within the first months of launch, advanced demand planning software not only offers improved forecasting, but a revenue and growth generator that promptly reacts to early market indicators and enables planners to prepare for the market reaction — ultimately reducing lost-sales and optimising stock.
