Predicting Tomorrow: Uses of Forecasting

Forecasting remains an important tool in the media, marketing and advertising analytics toolbox. We use forecasting models to predict consumer behavior, market shifts, competitive activity and shifts in demand. Our forecasting and scenario planning tools help our companies reduce risk and respond quickly to changing market dynamics.  And of course, we use forecasting models to predict uptake of new products and services – including media products and shows.

Yet forecasters face major challenges. Data sparsity is a growing problem due to privacy protections, distribution disruptions and signal deprecation. A large part of forecasting in marketing is trying to predict what things we will like (e.g., personalization, recommendations); but doing this right is hard enough when the data are solid. When the data signals are weaker, forecasts are weaker. Some companies have been doing work on this — aggregating individual-level data to become “cohorts” with similar tastes. The success of these approaches remains to be seen. But a big question hanging over all of it is whether consumers will still be offended and feel violated when the forecast is correctly made from privacy compliant sources. Can AI and “synthetic” modeled data offer an opportunity to overcome data limitations and enhance model performance?

At FORECASTING 2024 on July 17, a variety of companies will share how they are building forecasting models despite data sparsity and other challenges to effective modeling. Learn from the leaders in this fascinating and important space. Note that this event will be virtual.