Two things were essential for this online food wholesaler that delivers daily orders to his customers.
On the one hand, the customer (the restaurant) can not miss out on any ingredients. If this is not in stock, you can lose this customer.
On the other hand, the client (the restaurant) needs his products to be delivered on time, so that he can serve lunch and dinner.
The management team believed that intelligent demand forecasting would have great influence on their day to day operations and solve two key problems: Volatile demand was one of the problems neccesary to be solved. Erratic planning of workers in the warehouse to process the orders was the second problem that needed serious attention. The combination of the two had proven to be a challenge for traditional forecasting methods, leading to inaccurate labor planning and potential unsatisfied customers.
The SYMSON team kicked off the project with a brainstorm to identify possible drivers that needed to be included in the forecasting model. The customer came up with several internal- and external data points that were relevant in their market and to their customer behaviour.
The project team of SYMSON was engaged to connect the customers data into the forecasting engine in SYMSON, with the goal of forecasting sales up to 4 weeks in advance. Starting with a small scope of the locations, the SYMSON team worked on including various data points such as the weather, promotions, (local) events and sales data.
Forecasting is a particularly powerful tool for ecommerce businesses because the space provided allows access to an abundance of data and data streams – fantastic for some of the data hungry algorithms in SYMSON. This case clearly shows that a combination of qualitive data, powerful technology and a strong project team can produce accuracte results that clients need to be succesfull.