Retail media has reached a point where measurement is table stakes. But what table stakes actually look like is changing fast.

With the launch of its latest measurement tool, Kroger Precision Marketing (KPM) is signaling its priorities for retail media measurement: Speed, incrementality, and clearer decision-making.

• PrecisionView 360 (PV 360) is a cross-media measurement solution that enables marketers to understand how individual channels or combinations of channels are moving the needle.
• Control groups isolate the real impact of media exposure, revealing incremental return on ad spend (ROAS) and campaign-level management gives visibility into household reach.

“Measurement is the thing that you rely on to make your next decision,” said Chandra DiGregorio, vice president of data science and research at KPM. “It’s the kind of thing that you want to fade into the background and take for granted. But in order for you to get to that point, you need to be confident in the results.”

Meet in the middle: KPM is positioning PV 360 between two measurement extremes.

On one end is traditional marketing mix modeling (MMM): Comprehensive and cross-channel, but often slow and expensive. On the other end are standard post-campaign reports: Fast, but frequently lacking incrementality and cross-channel context.

PV 360 aims to sit in between.

“This product is really delivering that middle ground,” said Brian Spencer, marketing director at KPM. “It’s faster than MMM, but way more accurate and useful for making intelligent decisions than traditional media post campaign reports.”

Speed is a central part of that positioning. Instead of months-long MMM cycles, KPM says it can deliver results in under two weeks after a campaign ends.

“Once that post period wraps, we're delivering results within a couple of weeks. I think right now we're under two weeks and delivering results,” said DiGregorio.

That timeline matters in a retail environment where budgets are reallocated continuously.

Modern models: At the core of PV 360 is household-level incrementality.

• KPM uses loyalty data covering roughly 95% of transactions, online and in-store, to link ad exposures to verified sales.
• The system relies on randomized controlled trials where possible and synthetic matched control groups where necessary.

Those channel-level incremental ROAS results are then fed into a Bayesian market-level model, which combines past performance with informed assumptions to produce more reliable predictions.

“Those household-level numbers from that first engine are kind of fed in as the prior,” said DiGregorio. “So the model is really well-informed and anchored on a truth set.”

This approach attempts to address a common criticism of traditional MMM: That it can reinforce historical channel assumptions and make it difficult to detect breakout performance.

• “It's very difficult for individual channels to break out of that status quo,” said DiGregorio. “What we're really focused on is making sure that our Bayesian MMM is informed by the exact experiment that happened.”
• In theory, this allows for more campaign-specific nuance within a cross-channel framework.

But marketers first need to set expectations around incrementality, which typically yields lower returns than traditional ROAS reporting.

“Educating on incrementality is an always-on kind of thing because it is a higher bar. Incremental ROAS numbers are not as high as traditional ROAS numbers,” said Spencer.

This was originally featured in the Commerce Media Weekly newsletter. For more marketing insights, statistics, and trends, subscribe here.