As Adobe Embarks On Co-Op Data Project, It Must Create Value For All

gerard-dataxuData-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Gerard Keating, executive vice president of products at DataXu.

Adobe has announced a new cross-device co-op data initiative that serves as an ambitious counter to Google and Facebook’s now infamous walled-garden data offerings. Adobe is considered one of a small group of companies with the capacity and infrastructure to consider building such a solution and, in my view, the company has done so with all the right intentions.

But while many are lauding this initiative for supporting the interests of advertisers, Adobe’s current iteration of a co-op data model may run into significant challenges when it actually comes down to large brands and media owners freely sharing their data.

Adobe’s co-op data model reminds me of two large co-op projects I played a part in. Unfortunately, for similar reasons to the challenges Adobe now faces, neither of them were successful.

Learning From The Past

Sometimes valuable lessons can be learned from past mistakes. I worked on my first co-op project while at Index Technologies. IBM led the project, which was called AD/Cycle. Index Technologies integrated with IBM’s federated database and contributed our clients’ data assets to the co-op “mother ship” for the data-sharing good of all.

I worked on my second co-op project while I was at Engage Technologies, a CMGI company, where we led and built a federated solution to create aggregated behavioral profiles for communal use in advertising and content management.

The value proposition for a data co-op is that it offers a more valuable and data-rich solution than any one contributor’s individual reach. Both projects garnered significant traction in the trade press of their respective industries, with journalists lauding the imminent value of the co-ops to both vendors and clients alike. Ultimately, both projects failed.

Why did an initiative so widely praised fall flat? It comes down to a question of value. Will all the contributors to the co-op add the same levels of value? And if smaller company contributions to the data co-op are limited but end up reaping sizable benefits from larger companies, is there really value in the deal for the big guys?

With both the IBM and Engage Technologies co-op initiatives, all of the small to medium-sized companies rushed to participate and contribute to the co-op. For them, the co-op represented an exciting chance to expand their reach and opportunity. But the big players – the companies that already owned the majority of the data in their given space – were reluctant to participate and go all-in. They struggled to see the positive value of the co-op. What did they gain by turning over their data? The answer was unclear.

Index Technologies struggled with IBM’s co-op value proposition. We shared a concern that our smaller and similar-sized competitors would gain access to our customers’ data in the proposed federated database. Any one of us would gain the potential to lure each other’s customers away by providing an easy migration path.

Eventually, both the IBM and Engage Technologies solutions became stuck in what Gartner refers to as the “Trough of Disillusionment.” Neither initiative ever climbed the “Slope of Enlightenment” to achieve the hoped-for end state. And given these firsthand experiences from my past, I confess to some concern that Adobe’s co-op solution may not gain as much traction as the industry hopes precisely due to the same reasons.

Delivering Value From Data To Large Companies

The concept of a co-op, in general, is strong. I was also fortunate enough to be involved as a vendor at Enpocket in a variant of a co-op solution that successfully enabled customers to monetize their extremely valuable corporate data asset for their exclusive benefit. It was a true win-win scenario for all.

At Enpocket, our team targeted large telecommunications companies that had profiled but not yet monetized their user bases. These companies owned large, deterministic data assets that combined both demographic and behavioral data.

Enpocket enabled these telcos, including Sprint, to import their data assets into our ad-serving solution and subsequently offer highly targeted opportunities to advertisers. This value proposition was strong. It resonated particularly well with these large carriers because it was a private, secure solution.

Unlike the IBM and Engage Technologies data co-ops, these big data owners reaped the value of their data assets and quickly climbed into Gartner’s “Plateau of Productivity” where they realized the full benefits of their investment.

Next Steps For Adobe

Adobe is headed in the right direction when it comes to breaking down the walls of impenetrable walled gardens, but there is bound to be resistance along the way when it comes to getting large data owners to contribute their assets to the proposed co-op.

My expectation is that Adobe will ultimately morph into a privatized version of the same technology but within its DMP solution, working directly with its larger customers to help them build cross-device assets. By creating private graphs that stitch together a wide variety of aliases to create a complete picture of users, for example, Adobe will make their vast swaths of information actionable. This avoids wasted media spend and ensures measurable ROI in today’s complex cross-device environment.

Large customers of Adobe will see this as an opportunity to safely and securely sell their private, proprietary graphs to advertisers that can leverage the customer information for their own purposes by partnering with programmatic platforms for targeted, insightful campaign execution.

Adobe is well positioned to digest, consolidate and reconcile vast amounts of traditionally siloed data, but the key to success for a data initiative is creating value for all.

Follow DataXu (@DataXu) and AdExchanger (@adexchanger) on Twitter.

1 Comment

  1. A whole slew of private, proprietary graphs that are different and specific to each individual publisher would be a fragmented and confusing experience for buyers that are increasingly looking for a unified view of a user across media. Third party probabilistic solutions offer a holistic view agnostic of inventory source with more scale than any individual publisher's graph could achieve, albeit at lower perceived accuracy. There is still a lot of work to be done in striking the right balance between a trusted level of accuracy and cross device scale that can be easily targeted and measured, and I don’t think private graphs are the answer.

    Reply

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>