Today’s column is written by Nancy Marzouk, founder and CEO at MediaWallah.
I’ve noticed a recurring sentiment in recent conversations with several large advertisers: “I don’t want to leverage a walled garden but there doesn’t seem to be a better option.”
Why the dilemma? It’s quite simple. Brands are struggling with trade-offs. Matching customers and engaged and anonymous prospects across devices and media channels at scale in real time continues to be an issue. As mobile grows, this becomes even more imperative.
There shouldn’t have to be a trade-off. Advertisers need another solution other than using walled gardens to establish user connectivity – and that solution sits with them, not vendors. Advertisers don’t need packaged ad tech as it relates to data. They need an a la carte menu so they can leverage functionality that fits their data needs.
Some brands have started this migration. Gartner analyst Laura McLellan recently predicted that by 2017 CMOs would spend more on IT than their counterpart CIOs. The biggest driving force behind this shift is the building of an intelligence layer to leverage a brand’s data, which is paramount to drive corporate revenue.
Data Connectivity And Walled Gardens
For most brands, connecting consumers across devices is a complicated endeavor. On average, only 30% of an advertiser’s CRM data files are matched to online cookies. And on average, just 30% to 60% of an advertiser’s online anonymous and engaged prospects can be linked across media touch points and devices.
When you think about how many marketing platforms an advertiser uses and the fact that cookies are still the primary identifier used to sync people, it’s no wonder that data connectivity is problematic.
The idea behind walled gardens is that they can allow for better cross-platform user syncing and matching of online and offline user data. There are typically two types of walled gardens. The first type has rich registration data to maintain user persistency through deterministic matching, such as with Facebook and Google. The second type offers a full digital marketing suite so an advertiser can have better cookie connectivity by limiting the number of platform syncs and probabilistic matching.
The problem with walled gardens is that regardless of how users are connected, an advertiser must replace some marketing tech that’s working and accept decent connectivity for a fraction of their user event stream data. They’ll also be unable to successfully match data outside the walled garden. None of these caveats make sense for an advertiser.
Most brands do have some kind of ID tracking already in place for their customers. These brands use online registration and CRM automation as a way to extend these IDs online. They simply need a solution to maintain match tables and a mechanism to drop their own identifiers, such as first-party cookie to maintain user connectivity. A very low percentage of people block first-party cookies – less than 5% – and many would allow for better user identification over third-party cookies, which are used by most walled gardens.
Any marketing-tech vendors that drop third-party cookies could drop a first-party cookie if advertisers have an infrastructure to support hosting the vendor’s complete code. It’s not typically done because this would allow for complete transparency, which is outside of many vendors’ comfort zone. It also might commoditize other parts of a vendor’s platform and diminish its value.
First-party cookies can also extend across an advertiser’s digital paid media and owned properties for better engagement and anonymous prospect connectivity but they don’t solve the problem completely. Advertisers would still need to use probabilistic matching to compensate for data leakage due to cookie deletion and fragmentation of connected device among engaged and anonymous prospects. This is another a la carte feature, which today is packaged within ad tech.
I think the puzzle pieces are available for advertisers to solve user connectivity but advertisers must establish a proper infrastructure to support it and figure out which key components are needed to build their own solution.