Multinational companies face many marketing and sales challenges associated with regional differences and demand forecasting. The huge number of products alongside the complexity of different markets make it hard to predict performance and understand the rapidly changing macroeconomic environment. With copious data sources from various channels and regions, it’s not only difficult to create strategic plans, but the Covid-19 pandemic changed consumer behavior, which still needs to be understood and accounted for. Experiencing similar challenges, a company in the FMCG market hired Eraneos to help them build an ambitious solution that was worthy of the company’s global presence and extensive reach.
We set out to activate the client’s data and create strategic plans by, firstly, building a Data Hub on Azure that centralized all information incoming from different channels, regions, and products while processing information uniformly. We also added machine learning components for more advanced analytics built on top.
By ensuring all of the data was fed into advanced dashboards that can be analyzed by country, region, market, and product levels, we created dynamic forecasting based on data such as GDP and market trends. This allowed it to forecast sales and marketing performance for product segments and markets within each country for up to three years ahead.
A leading global consumer goods company. The client is active in several consumer goods categories in more than 70 countries across six continents.
The dynamic forecasting solution has advanced dashboards that allow the company to break down performance at a granular level. The client is also able to create strategic plans within every region for its hygiene, health, and nutrition markets.
Powered by an Azure Data Hub that ingests all of the data from different channels and sources, the solution creates a forecasting model for each product, every month. As a result, more than 30,000 forecasting models are now created at the company automatically on a monthly basis. This means it is not only able to plan strategically for the future, but predict well in advance any unexpected challenges that might occur.
To counter the sudden effects of consumer behavior due to Covid-19, we added different scenario planning options to the forecasting models such as short-term pantry stocking, rapid stabilization, and dip and rebound so that the model takes into account the sudden spike in the demand for some products. Equipping our models with a Covid impact regressor allowed us to deliver very high accuracy during these months of uncertainty and adjust for demand in the future. Because we already had a strong foundation in the Data Hub, we managed to deliver everything within 2 months, allowing the client to react quickly to these changes. With our dynamic forecasting solution, they are able to predict demand well in advance and plan strategically for the future.