What will happen tomorrow, next quarter, next year?
- When will the customer buy the bicycle—two weeks after they see the Facebook ad? Does the weather impact sales and, if so, how much?
- How will the market react to the TV spot? Will viewers visit the landing page or call the toll-free number—and why?
- What can call center data tell us about which customers will churn if we raise annual fees?
- Overall... how can we respond to continual and turbulent changes in the marketplace?
Use forecasting and predictive modeling to answer these questions. Apply the learning to your media mix and watch your marketing ROI skyrocket.
Our process for forecasting and modeling includes:
- Sales forecasts and media/marketing mix modeling
- Statistical forecasts of time series data, including trends, cycles and seasonality, and abrupt shocks
- Includes the “de-composition” of top-line sales into major components—baseline vs. different media channels (TV, etc.) and other effects, such as pricing
- CRM models
- Response, retention/defection, cross-sell, LTV, share of wallet models and price elasticity models, as well as website visitor intent
- Includes “scoring code” for deployment in tactical targeting
- Attribution models
- Analysis of omni-channel customer contacts, and their relative impact on KPIs
- Includes digital contacts, as well as offline, mass and other influencers, such as offer
- Anomaly detection models
- Development and deployment of predictive models that identify important anomalies and outliers in data
- Includes fraud detection, ghost users and key opportunity moments, such as impending purchase