Predicting B2B Demand for Highly Configurable Products

High Tech Industry

It’s famously difficult to predict demand in the dynamic and complex high-tech industry. The challenges arise from a highly dynamic market environment with rapid technological changes, diverse customer segments, and multiple distribution channels. The configuration of the products will change during the sales cycles, while key components have long lead times, forcing the companies to plan their components before the final configuration is known. Additional complexity is added when considering that it’s not yet clear when the demand will materialize throughout the sales cycle.  

Yet, some high-tech companies can thrive in this difficult environment. We recently met with one of o9’s high-tech clients and discussed their journey of resolving the challenges of accurately predicting the demand of their highly configurable products.

Use Case

The company is managing an order book that is almost completely configure-to-order, where 90% of their configurations never repeat. How do you plan your supply chain activities when your pipeline constantly changes due to customers asking for new configurations, new delivery dates, or different quantities? 

Inaccurate planning led to high inventory levels, excess and obsolete write-offs, and part shortages. They had previously struggled to break the silos between CRM sales planning and supply chain planning, as the data in the CRM couldn’t be leveraged by the supply chain due to low data quality and missing attributes needed for planning purposes.

The resolution was to leverage o9’s flexibility and incorporate the available data from the CRM and enhance it in o9 to make it useful for planning. o9 became the system of record, and enabled the following three key solutions to improve the value leakage:

• AI/ML for analyzing deal predictability: Likelihood of winning a deal is forecasted based on deal characteristics and historical data. o9 calculates deal uncertainty which helps planners focus on these deals to save time and increase value-adding activities. This is performed using daily snapshots of the full order book and pipeline, monitoring how orders are changed across quantities, products, and requested delivery dates. 
• Account-based forecasting: The top 20-30 accounts are reviewed to ensure that the demand is correct and that there are no duplications of forecasts. 
• Attach-rate forecasting: Is used to generate a top-down component level forecast. o9 uses an attach-rate forecast with probability of a certain configuration on each layer of the planning BOM, up to 8-10 layers deep, which improves the accuracy of the top-down component level forecast.

Read more about this and other critical use cases for driving resilience in the High-Tech Industry in our White Paper “Critical Use Cases for Driving Resilience in the High-Tech Industry” by o9 solutions.

Future-proof business solutions with GenAI

Supercharge planning and decisions with GenAI’s intelligent insights. See how it transforms your business.

Request a demo