Measurement

Goals and KPIs

Understanding the impact and ROI being delivered by marketing channels and campaigns may be both the most important and the most difficult aspect of performance marketing.

Marketers must first clearly define the goals of campaigns, starting with the top-level business goals and related marketing objectives that each campaign is intended to help drive. It’s essential to then identify a single primary KPI, as well as any number of secondary KPIs, which will drive short-term performance reporting and optimization decisions, including automated bidding.

This KPI decision is more crucial than is often realized and both audience targeting and user experience must be aligned with it in order to drive impact efficiently.

Getting 'less wrong' all the time

Beyond determining each campaign’s KPIs, marketing organizations must also ensure that they are able to effectively measure the actual impact of their campaigns on those metrics.

Unfortunately, there is no way to ensure 100% accuracy in measuring the true business impact of marketing investments - it's just not possible. But that doesn't make it any less crucial to build and continually refine measurement strategies that minimize those inevitable inaccuracies.

To quote the wisest advice I've heard around marketing measurement: "There's no way to be perfect, it's just about being less and less wrong every day, every quarter."

2 primary measurement methods

Many companies today are still far too reliant upon attribution modeling as their primary method of measurement, despite the ongoing degradation of user-level tracking capabilities that continues to make attribution models less and less accurate.

Attribution modeling is still absolutely necessary as one component of effective measurement, since no other method is sufficiently granular or real-time to drive automated bidding and in-flight optimization decisions. However, in order to ensure positive ROI, attribution model outputs must be calibrated at the channel or sub-channel level using more accurate (though less real-time and less granular) measurement methods, particularly incrementality.

For most companies attribution and incrementality together are sufficient for effective measurement, but many enterprise businesses can also benefit from leveraging marketing-/media-mix modeling to provide periodic estimates of the impact of marketing investments across all channels. However, again, MMM outputs should ideally be calibrated at the channel or sub-channel level using incrementality test results, in order to improve the accuracy of the insights they provide.

Google Sites

Report abuse