GeoLift by Recast is built to help you design and analyze rigorous geographic incrementality experiments. However, when using GeoLift or any other similar tools there’s one issue you should always be aware of: the problem of “generalizability”. The problem of “generalizability” plagues all types of experimental design – researchers and […]
Considerations When Designing GeoLift Tests Read More »
What are probabilistic forecasts? We care about two things when we’re making forecasts: how “correct” is the forecast (how far away were the actuals from the predicted values) and how good are we at understanding our own uncertainty (we want to penalize a forecast less if it was said to
Using CRPS to Evaluate Recast Models Read More »
Designing incrementality experiments can be tricky for businesses that have substantial amounts of natural or induced seasonality in their business. Big events like black friday or cyber monday discounts can be make-or-break for certain businesses and so it can be critical to understand the incrementality of advertising during this time
Designing Incrementality Experiments During the Holidays Read More »
It’s tempting to evaluate a channel by its average ROI. If you spend $100,000 on Facebook and generate $200,000 in revenue, that 2× return looks great. But average ROI is a backward‑looking measure. It tells you how the last $100,000 performed — not how the next dollar will perform. To
Average vs marginal ROI in MMM: when to use each (and why both matter for your budget) Read More »
How do you make sure your model doesn’t spit out negative ROIs on channels that are clearly working? Or attribute three times more revenue than your business actually drove? Or collapse when two correlated channels move together? These are the kinds of problems priors are meant to solve. They’re one
What Are Priors in MMM – And Why They’re Difficult to Get Right (But You Need To) Read More »
The board just asked for ROI improvements by next quarter. Your growth is flat. Pressure is rising. Budgets are under review. Marketing teams tend to go one of two ways here. Some spiral into what we call the Inaction and Panic Cycle. Others take control through a system of Action
The Inaction and Panic Cycle: How Poor Measurement Traps Marketing Teams Read More »
Media mix modeling introduces statistical rigor, causal inference, and a more accurate picture of what’s driving growth across your marketing channels. But it also introduces uncertainty — and if that measurement uncertainty isn’t framed correctly, it can feel like confusion rather than clarity. That’s why so many MMM efforts stall
How to Talk About Uncertainty in MMM Without Losing Executive Trust Read More »
More and more brands are running lift tests. That’s a good thing. Incrementality experimentation methods are not that complicated — and you don’t want them to be. The value of a structured experiment is that interpreting the results should be straightforward. But in practice, we see some common ways teams
The Most Common Mistakes Marketing Teams Make when Interpreting Lift Test Results Read More »
Attribution systems break at scale because they can’t measure causality. As businesses grow more complex, they often fall into what we call the ‘inaction and panic cycle.’ This article explains why attribution fails – and how to replace it with an incrementality system built for modern growth. Attribution is not
Why Attribution Breaks at Scale (And What to Do Instead) Read More »
Many businesses face tough challenges when it comes to measuring marketing incrementality because the nature of their product implies conversions that happen at multiple stages. Marketing activity drives new users or customers to the first stage, and then some proportion of those customers “convert” into a subsequent stage. This may
Product Update: Multi-Stage Models Read More »
