Successful Consent Marketing in Times of Data Privacy: Campaign Optimization With Marketing Mix Modeling

With third-party cookies, it is still possible for advertisers to track and analyze user behavior via their ads on other websites and to create user profiles based on this data. This allows users to be presented with targeted ads tailored to their preferences. Microsoft, Firefox, and Apple have put a stop to this, and Google will soon follow: Starting in the second half of 2024, the company wants to prevent cross-website tracking through third-party cookies in its own Chrome browser and replace it with privacy-compliant alternatives (“the privacy sandbox”).

Browser and operating system providers are responding to the growing interest among users in topics such as data protection and transparency. Added to this are changes in the legal framework and considerations about the positioning of their own solutions and platforms.

We leave the interpretation of the revised Data Protection Act, which comes into force in Switzerland on September 1, to the lawyers. Regardless of this, we are seeing that many companies are no longer only oriented toward what is legally necessary, but increasingly want to respect their customers’ need for transparency and consent to data processing, for example, with opt-in cookie banners.

User identifiers, which are often stored in cookies, attribute interactions with advertising media to success signals on a website. Ad blockers or corresponding browser settings have always ensured that only a part of the interactions can be measured and the data is incomplete. However, the current discourse around consent marketing has made our customers increasingly aware of this issue and raised the question: How can we properly classify the measurement of success in digital marketing?

To be able to carry out significant marketing analyses in the future, additional methods can be applied.

Many advertising platforms therefore already rely—sometimes more transparently, sometimes less—on modeled figures. This keeps complexity low, but increasingly raises the questions: Who should be in control of such projections, and how do we properly classify the different marketing channels?

Webrepublic recommends combining different approaches to evaluate marketing success holistically. For example, controlled experiments can be set up using regional targeting to determine the success of a marketing channel or campaign in this region and compare it with others.

Such insights serve as input for comprehensive marketing mix modeling. This helps to measure, optimize, and plan investments. Past and future effects of marketing measures on defined business objectives are examined and compared with each other. The results of these two approaches can be used to manually adjust digital attribution.

Webrepublic supports clients in the implementation of consent management platforms. Part of such a project is to find the optimal setup for the customer and to support them in transferring this into their daily marketing routine.

Often, implementing consent management platforms is about how comparability can be achieved. Most modern tools provide information about how many people consented in aggregate, but not about user identifiers and user segments. To obtain this information, Webrepublic uses approaches from marketing mix modeling to provide an informed estimate of how the measured data would have evolved if no consent banner had been introduced (“causal impact analysis”).

This approach allows us to set correction factors that make campaign successes before and after the introduction of consent management more comparable.

Data-driven digital marketing is becoming more challenging. Success measurement is moving from the actually measured (“digital attribution”) to modeled data. Whether to entrust this modeling to only advertising platforms depends, among other things, on the individual advertising mix, the digital and physical customer touchpoints involved, and a cost-benefit calculation.

Clean and well-structured data will become increasingly important and will decisively contribute to the success of analyses and future campaign planning. Experiences from Webrepublic show that in marketing mix modeling projects, a large part of the time is needed for gathering and categorizing data. It is worthwhile to rely on a clear naming convention as early as the campaign setup stage and to define an overarching data architecture and to transfer this to continuous, preferably cross-channel, reporting and monitoring.

Even if advertisers must make do with less information, good, data-driven digital marketing can still be implemented.