Consumer goods manufacturers are striving to meet the demands of their customers. They are keen to have a robust and stable approach to the category and segment forecasts, installing a stronger confidence in the predictions. Category usually refers to the entire product whereas segments are the variant forms under which the product is manufactured.

Transactional data can be merged with other data sources and converted to desired form suitable for analysis. Multivariate time series (ex. Vector Auto Regression (VAR) model) can be used to identify the key drivers in the category and segments to be forecasted.

Univariate analysis can also be done, if required. Additionally, Stationary test can be done and differencing of the series can be made whenever required. Seasonal variation (festival, unexpected weather) can be accommodated in the model.

Dexterity provides volume forecasts using Time series models. Intuitive dashboards are shared to identify the changes in the forecasts in the presence of marketing events.

An output with 3 layers can be provided:

- Diagnostic part: Key drivers for the current segment evolutions

- Forecasting part: Forecasts for category and segment sales

- Simulation part: A tool reflecting the changes in forecast when some of the input variables change

Talk to one of our Marketing Analytics Experts today to learn how we can help you in decision path modeling.