• by Laurie Sullivan , Staff Writer, Yesterday

Google has made changes to its marketing mix modeling (MMM) tool called Meridian, giving marketers more visibility into the data.

The platform now allows marketers to include non-media variables like pricing and promotions to precisely measure their impact on sales.

To help guide the MMM with business data, Google has introduced new channel-level contribution prior to delivering more relevant insights. 

For example, if a marketer has an intuition about the proportion of the key performance indicator (KPI) that could be incremental because of a particular channel being used like video, the marketer can set a "prior" on the contribution of a given channel by setting certain priors.

A "prior" for a "custom incremental KPI per cost" (IKPC) is a statistical parameter that helps define the expected range of a marketing channel's incremental contribution to a KPI for a specific cost level. It is particularly useful in MMM when dealing with complex outcomes or when standard priors are insufficient, according to a Google post.

Marketers can also now measure the longer-term effects of upper-funnel media, considering the brand recall and influence that drives a purchase weeks later.

This is made possible by enhanced binomial adstock decay functions, a model used in MMM to represent how the effectiveness of advertising spend builds and then decays over time.

The adstock decay functions are “useful for modeling media channels that have persistent, long-lasting effects, such as TV and radio, where a significant portion of the impact can occur in the latter half of the effect window for smarter optimization,” according to Google.

There are 30 new certified global partners and an active Discord Community for marketers to get expert help to deploy Meridian's capabilities and translate insights into growth. 

The timing is interesting. Recast, an incrementality measurement and forecasting company, also released an update to its platform today. 

The launch of “GeoLift by Recast” provides a stand-alone incrementality platform to determine what the company calls “true” cause or impact of marketing on a business. 

Marketers using Recast's product can explore different experimental designs, understand historical variance, and run experiments without relying on outside consultants or data-science teams.

The platform also helps validate MMMs and feed model refreshes with reliable calibration data.