Upwave is moving a new approach to brand-lift measurement into closed beta, introducing Predictive Brand Lift, a surveyless methodology that uses Bayesian modeling and AI to estimate advertising impact.
The measurement company said the product has been developed and tested for more than a year with selected customers. The closed beta will expand testing across additional campaign types, industries and media environments before any broader release.
Traditional brand-lift studies typically compare exposed and control groups using consumer surveys to determine whether advertising changes measures such as awareness, consideration or purchase intent. The Interactive Advertising Bureau defines brand lift as the change in consumer perceptions associated with exposure to advertising, commonly measured across metrics including awareness, familiarity, favorability and purchase intent.
Upwave’s Predictive Brand Lift takes a different route. According to the company, its system applies proprietary Bayesian modeling and AI to historical Observed Brand Lift data to estimate brand lift without fielding consumer surveys. Upwave says that dataset includes measurement experience across more than 3,000 brands.
The company is positioning the predictive methodology alongside, rather than as a replacement for, its existing Observed Brand Lift product. The latter uses survey-based measurement and analytics to directly assess campaign impact, while Predictive Brand Lift is intended for campaigns where survey-based measurement may be less practical.
That distinction is important for advertisers. Survey-based brand studies provide direct consumer responses, but they can require sufficient audience scale, time and research infrastructure. A predictive approach could potentially make brand-effect measurement more accessible across smaller or more fragmented campaigns. However, the usefulness of such a system ultimately depends on the quality of its underlying data, model validation and the degree to which predictions correspond with observed consumer responses.
Upwave has not publicly disclosed enough information in the announcement to independently assess the methodology’s accuracy, model inputs, validation results or performance across different media environments. Those details will become particularly relevant as the company expands the beta.
The development arrives as the advertising industry continues to seek more comparable and scalable approaches to measurement. IAB’s Measurement Center identifies fragmentation and inconsistency in advertising measurement as ongoing industry problems and is working on frameworks covering exposure, reach, attention, attribution and outcomes.
The industry is also increasingly distinguishing between measurement methods rather than treating a single metric as sufficient. IAB and the Media Rating Council’s attention framework, for example, separates different approaches to measuring attention and emphasizes methodological consistency and accountability.
For agencies and advertisers, Upwave’s two-method model could create a practical choice between directly observed survey-based measurement and model-based estimation. But predictive measurement also raises a familiar question in advertising analytics: how much methodological transparency is necessary for buyers to trust modeled outcomes?
The closed beta therefore represents less a finished alternative to conventional brand studies than a test of whether historical measurement data and statistical modeling can extend brand-lift analysis to more campaigns. Upwave’s next phase will provide an opportunity to demonstrate whether its approach can maintain credibility as the range of advertisers, media environments and campaign conditions expands.
Market Landscape
Brand-lift measurement has traditionally relied heavily on exposed-versus-control research and consumer surveys. Standard measures include awareness, ad awareness, message association, familiarity, favorability, consideration and purchase intent.
The broader measurement market is moving toward greater standardization and comparability. In August 2026, IAB finalized Campaign Data Standards 1.0, designed to create a common framework for describing advertising experiences and inventory across platforms, channels and campaign systems.
That environment creates demand for measurement approaches that can work across fragmented campaigns while maintaining methodological transparency.
Strategic Outlook
For advertisers, surveyless measurement could potentially lower the operational burden associated with measuring campaigns that do not fit conventional brand-study requirements.
For agencies, a predictive approach could provide another measurement option when campaign scale, timing or audience characteristics make traditional surveys difficult.
For measurement providers, however, predictive methodologies create a validation challenge. Buyers need to understand how models are trained, what inputs influence estimates and how predicted lift compares with observed results.
Upwave’s decision to run a closed beta is therefore significant because the next stage is about validation rather than simply product availability. The company’s ability to demonstrate consistency across different campaign environments will determine how useful predictive brand lift becomes as an alternative measurement methodology.
Top Insights
- Upwave has placed Predictive Brand Lift into closed beta.
- The methodology uses proprietary Bayesian modeling and AI, according to the company.
- Predictive Brand Lift is designed to estimate brand impact without consumer surveys.
- Upwave will continue validating the approach across additional campaigns and media environments.
- Methodological transparency and validation will be central to advertiser adoption.
Get in touch with our Adtech experts.




