When building out MNTN’s incrementality reporting, we evaluated methodologies used across the industry, weighed their pros and cons, and identified areas in which we could improve the accuracy of measurement by removing potential data biases. From this approach, MNTN selected control vs. exposed testing, which results in accurate and actionable data without requiring any upfront costs or planning.
Calculating incrementality is simple. First, we define your exposed and control groups.
- Exposed Group: These are users who have seen your brand’s ad.
- Control Group: A control group who did not receive your brand’s ad is generated from remaining users who match your prospecting audience’s attributes. To allow for an apples-to-apples comparison, the control group is randomized with statistical significance to mirror the volume of the exposed group.
Next, we subtract the conversions and revenue of the control group from the exposed group to provide a clear view of the incremental lift in outcomes driven by MNTN prospecting.
Three reasons we use this approach to calculate incrementality:
- Efficiency: Comparing the results from exposed groups against the results from randomized control groups removes the need for upfront planning and costs that result from traditional forms of incrementality measurement, such as placebo testing or geo testing.
- Accuracy: MNTN’s exposed vs. control testing isolates key variables that could otherwise skew the results that would have occurred without MNTN prospecting. That’s why we:
- Compare results from users that match the attributes selected for your prospecting audience, which means the only variable that differs is exposure to your ad.
- Don’t rely on geos, which can introduce additional variables to the test that could impact outcomes, such as regional cultural influences.
- Neutrality: Similar to Meta and other platforms that use a similar approach, MNTN randomizes our control group to remove the potential for unintended biases