Online R&F: Combining Panel and Ad-Serving Data
On Monday, Atlas DMT released the results of a study that tested the approach, as well as a new tool - the Atlas GRP and Reach Forecaster - that uses the combined methodology for predicting online GRPs, TRPs, reach, frequency and marginal cost for Internet media buys.
For the study, Atlas DMT used its Atlas GRP and Reach Forecaster to estimate the reach for 327 online media buys of varying duration across 77 different publisher sites. The media buys took place over a six-month period. Atlas DMT then calculated the actual reach of the buys for comparison by using the Atlas Digital Marketing Suite's back-end reach and frequency reports. Using a standard statistical analysis for calculating the correlation between pairs of pre- and post-campaign reach numbers, evaluated both the forecasted and actual reach figures for each media buy.
The study results indicate that the Atlas GRP and Reach Forecaster predictions were highly correlated to the actual campaign reach; capturing 83% of the variance in reach for all 327 online buys. By using the combination of user and server-centric data, the Atlas GRP and Reach Forecaster is able to predict directional reach estimates on the front-end of a campaign before a media planner has spent one dollar online. By using third-party ad serving technology in combination with panel data, media planners are able to obtain granular reach and frequency analysis at the site, creative and campaign level.
"Until the introduction of the combined approach, the industry had to rely on syndicated panel data and guesswork to make strategic decisions," said Young-Bean Song, Atlas DMT's director of analytics and the Atlas Institute. "Media planners can use the Atlas GRP and Reach Forecaster to allocate budget across websites, and balance maximum reach with effective frequency levels. Furthermore, the exciting part about this research is any advertiser using the Atlas Suite for third party ad serving can recreate this exact study with their online campaigns."