Ad-Tech, and the agencies they serve, graduated to an ‘audience-based’ methodology because of the higher degree of transparency it offered in targeting first and third party offline and CRM company lists. Until then the Ad-Tech infrastructure has largely gotten away with transacting real-time bidding against inferred audience characteristics defined by site visits and behind rather opaque claims of reach & accuracy.
The true beauty of onboarding is that it offers a method for pushing audience targeting and attribution into the open. However, as audience-based onboarding has matured so has the marketing spin. While we strongly support the value of onboarding as a means to get more transparency and accountability into the campaign process, we also believe that agencies and their clients should be informed of where the marketing may not be perfectly aligned with the messaging.
At the top of the list are the limits on the individuals actually being matched to. The now popular “person-based matching” claim still requires matching on emails or social IDs that have been self-reported at known household addresses. The addresses provide demographic and psychographic color to the audience. Offline household files are rigorously compiled and double and triple verified for accuracy of the postal address and the head of household. Self-reported emails and social ID addresses? Not so much.
A systemic limitation of offline compiled data is that it is limited to adults with a commercial footprint, such as credit, name on a magazine subscription, or a voter at home. This is a good thing for privacy and accuracy; however, multi-verified sourcing often results in people and homes not being covered because of outdated and incomplete information – especially when it comes to matching to all of the individuals that may be online across multiple devices in a home. There are no sources for a business email-matching equivalent at scale.
The cookie-based onboarding companies still largely rely on third party email sources as a match key. Their sourcing of emails is broad and excessively duplicative. A person and their reported address can be extremely inaccurate or purposely misleading. An email associated with one person is often associated to multiple identities, devices and postal addresses; and can include business emails, former addresses, fake addresses, and phony naming conventions. Dozens of emails per person filling up the match rate report creates an email ID mess with many false matches, especially when matched to social media and generic free email providers like Earthlink, Hotmail, Roadrunner, Yahoo and AOL.
Active cookie coverage on browsers is the choke point in the onboarding process. A large percentage of the marketplace (especially business networks) block third party cookies. Onboarding a CRM file to target businesses is a challenge even with cross-device matching. Google and Facebook IDs address some of this problem, but primarily for the benefit of their own display inventory and not for business use. The other challenge is mobile devices that don’t support third party cookies. Mobile targeting has to be device-ID or IP-based in order to address mobile activity. In a marketplace that is now more mobile than desktop, this is required.
Finally, the match rate claim for onboarding of a first party or third party offline file has crept up to around 50% except in local market trade areas, where it can be less than half that depending on the distribution of people and device types. There still will be, however, extensive duplication at the person and household level within onboarded audiences, as the cookies will be creating many matches for each user.
The alternative method for onboarding is device and IP location matching that is able to identify all of the devices within a household – and map each device to both a household and a person based ID. Devices and IP address are associated with a household, a business, a person ID, gender designation, an occupation AND the demographics associated with the individual.
IP and device locations create a persistent match key that can be used before and after a campaign to enhance attribution.
The IP and device location method allows for better coverage in the initial matching, informs our ability to execute deterministic attribution, and significantly reduces waste from the duplication of users within the campaign audiences.
Contact us today to learn more.