Leveraging Big Data to Make Cross-Channel Value Measurement a Reality
By calculating how much of an impact it has on keyword performance, the models can measure the display ad's cross-channel value per impression (VPI). The ad's VPI then becomes a key input into setting the appropriate bid and overall allocation for a given display ad. Armed with this knowledge, advertisers can go from investing and bidding based on instinct to instead bidding each ad based on its cross-channel value.
As with big data in general, this type of granular analysis, which would be cumbersome if not impossible for an analyst, becomes eminently doable with automated and self-learning models that can measure value over time. Moreover, it can be applied across millions of potential relationships between ads and keywords to uncover the atomic-level, cross-channel relationships that, in sum, drive big value to the advertiser's bottom line.
The same approach can be used to measure the cross-channel value of other traditionally hard-to-measure channels, such as Facebook or Twitter. It can even go the next step to optimizing this data across channels, orchestrating each individual ad and keyword to maximize cross-channel performance and get advertisers to their chosen overarching financial goal.
In the midst of much uncertainty around cookies and continued measurement challenges, digital advertisers can take heart in knowing that there are new, alternative ways to get to cross-channel value measurement that render the cookie debate irrelevant. Indeed, advanced cross-channel modeling at the atomic level taps the big data advertisers already have and conquers the issues with cookies to (finally) get advertisers to the goal of cross-channel, cross-device value measurement and optimization. All of this leads to better pricing and ad allocation decisions and, ultimately, a truly optimized digital media mix.
Robert Cooley is the chief technology officer at OptiMine, a bid optimization software provider.