Model transparency and accountability are foundational to our work at Recast. We openly publish model performance metrics to customers and use them internally to assess and improve the quality of Recast’s media mix models. Three of these metrics – Median Holdout Accuracy, Marketing Dollars Saved Using Recast, and Average Annual Value per Recast Model – are published on our live accuracy dashboards. This article explains what these performance metrics are, how they’re calculated, and why they’re important to our customers. Metric 1: Median Holdout Accuracy Simply stated:Median Holdout Accuracy – or Forecast Accuracy – is calculated as 1 minus the scaled Continuous Ranked Probability Score (CRPS) on the cumulative predictions and actuals over the forecast period. How it works in…
One of the ever-present problems with marketing mix modeling is that you always have to choose some start date. And since you always need to choose a start date, there’s always some period before the start date that’s impacting your results. Here's how Recast handles these carry-over effects.
Mockingbird empowers parents with premium, well-designed baby gear like their signature Single-to-Double Stroller. With a complex buyer’s journey and an…
With digital tracking breaking, consumer brands that relied on multi-touch attribution are now looking for alternatives to measure marketing effectiveness.…
There are three main ways that consumer brands measure marketing effectiveness: digital tracking (MTA), marketing mix modeling (MMM), and testing/conversion…
Business environments are messy; people with different responsibilities need to work together on decisions quickly and without perfect information. That…
In the context of marketing measurement, marketing analysts and marketing scientists use the term “incrementality” to refer to causality. Incrementality…