Measuring ads amid walled gardens and the open web

“Measurement is the fuel that powers smart decision-making,” our analyst Evelyn Mitchell-Wolf said during a recent webinar. As advertisers navigate the rise of walled gardens and the challenges of the open web, leveraging a multi-layered measurement stack becomes essential for building successful campaigns.

Ad measurement challenges across ecosystems

The current ad ecosystem presents unique measurement hurdles due to privacy changes and signal loss. In walled gardens, where channels like connected TV, social media, and retail media rely heavily on first-party data, advertisers benefit from closed-loop attribution and logged-in audiences.

However, Mitchell-Wolf pointed out that, “Platforms often limit access to user-level ad performance delivery data due to privacy requirements or their own financial interests.” These restrictions make attribution across platforms more difficult and leave advertisers struggling with probabilistic approaches to ad measurement.

On the open web, the challenges are more pronounced as logged-out audiences dominate and reliance on third-party cookies remains a sticking point. Less than 10% (8%) of advertisers said they faced attribution and measurement complexity in walled gardens, while 30% said the same for open web efforts, according to a Digiday and PubMatic survey.

While Mitchell-Wolf said no single measurement tool offers the granularity needed to satisfy every function, a multi-layered stack can help.

A comprehensive measurement stack includes three layers:

Top level: Media mix modeling

Media mix modeling (MMM) provides a high-level view of which channels drive overall business outcomes. “MMM paints the big picture,” Mitchell-Wolf noted, by assessing and predicting the effect paid media has on a brand’s sales. MMM also respects user privacy because it doesn’t rely on user-level data.

MMM can be resource-intensive and lacks the granularity needed for tactical decisions. Still, advances in automation and AI-powered tools are making MMM faster and more accessible. Meta and Google now offer their own open-source MMMs, Robyn and Meridian, respectively, further democratizing access.

Middle level: Incrementality

Incrementality testing enables advertisers to determine how different platforms contribute to performance. “Incrementality can assess the influence of media on brand lift, metrics like awareness and consideration, customer lifetime value, and more,” Mitchell-Wolf explained. However, incrementality is resource-intensive like MMM, and its lack of standardization can complicate cross-platform comparisons.

Despite these challenges, incrementality lift testing remains versatile and is increasingly popular. Over half (52.8%) of US advertisers plan to add incrementality testing to their measurement strategies, according to a Snap and EMARKETER survey. Incrementality is particularly valuable for optimizing platform mixes within a channel.

Lowest level: Platform reporting

At the most granular level, platform-specific reporting helps advertisers fine-tune creative and format choices. After web analytics (77.0%), platform reporting (61.2%) is the second-most used type of conversion measurement, according to the Snap and EMARKETER survey.

While easy to access, cost-effective, and indispensable when optimizing for a single platform, its reliability is often questioned. “Self-attribution is not always trustworthy,” Mitchell-Wolf said, and it’s not always comparable across platforms.

Watch the full webinar.

This was originally featured in the EMARKETER Daily newsletter. For more marketing insights, statistics, and trends, subscribe here.