Guest Blogs

Data Collaboration Is Getting Harder – Time for an OpenAPI for Data

Navigate the data landscape with precision – discover how companies leverage big cloud data giants such as Snowflake and AWS for seamless storage.
Data Collaboration

Just about ten years ago, the OpenAPI spec became a hot topic for programmatic advertisers. Today we use the term “interoperability” to talk about connecting pipes across the digital advertising ecosystem, but back then, we simply talked about the fact that everyone was drowning in complexity. Publishers, brands and agencies had too few technical resources to create custom integrations, and the work it took to configure each connection was slowing down our adoption of new capabilities.

OpenAPI became a rallying point to encourage more companies to make it easy to integrate.

Like so many things in life, the digital advertising ecosystem is cyclical, and we’re drown in complexity again. This time, in data.

More Data than You Can Possibly Imagine

By one estimate, we create 328.77 terabytes of data per day. A lot of that data flows through advertiser, publisher and tech company websites, data centers and data bases. From Google and Amazon to Experian, Chase and Walmart. As marketers have become wise to the value of their own first party data, they’ve started enriching first party data with a huge volume of insights from consumer shopping and browsing behavior, demographics, transactions, weather, location and more. The data comes in a huge variety of formats, from mobile IDs to offline PII.

With the data explosion has come two different approaches to housing data. A large number of companies, usually a bit less technical, use the big cloud data companies like Snowflake and AWS. Other companies that want a bit more technical control work with independents like InfoSum.

This alone has already created various walled gardens of data that need to talk to each other in order to support identity-driven advertising and marketing, which makes it hard for publishers in particular, who need to be able to work with everybody.

It’s about to get worse. As more companies get wise to the value of first party data and lean into identity-based advertising, they’ll start investing more in a data solution that they can manage more tightly. Particularly financial services and other companies with data that needs encrypting. That will give rise to a new category of data management where data never leaves a company’s walls and is encrypted on site.

We’re no longer talking about walled gardens, we’re now talking about thousands of data islands (including the walled gardens or “data gardens.”. And that is exactly what we don’t want, because it makes the whole idea of using first party data in an interoperable way nearly impossible – especially for partners that are under-resourced like publishers.

Let’s Work Together

A few years ago, I would have argued that Snowflake, AWS and Google need to talk to each other more easily, and that the independents need to designate resources to integrations with these big guys to make data collaboration easier. The world has evolved, and that won’t be enough to solve our data collaboration woes.

What’s becoming clear is that we need an open standard that enables interoperability and data collaboration.

That standard needs to:

Work across a huge variety of different scenarios – data clouds, independents, and in-house solutions
Be robust enough to support real-time data management
Allow for different levels of encryption and data security
Enable data owners to have transparency around precision and scale, and dial these levers up or down as needed

Standards Are Good, But Not Enough

Earlier this year, the IAB Tech Lab launched an initiative to provide standards for data clean rooms but this is just one piece of the puzzle. Having data clean room standards will help publishers with some level of integration efficiency. Ideally, after the first integration is complete, subsequent integrations with the same clean room should be a much lighter lift.

Integration standards to solve for interoperability, however. As long as data gardens are allowed to exist, a buyer will not be able to get a truly unified picture of their media buy, which makes things like attribution difficult – leaving us worse off than the heyday of third party cookies. While there is some pressure in the EU for the walled gardens to be more forthcoming about data transparency, interoperability is still a long way off.

For more such updates, follow us on Google News Martech News


Nancy Marzouk, CEO & Founder of MediaWallah

Nancy Marzouk believes that the future of customer experience lies with giving marketers, publishers, and platforms the power to connect data on their own terms. In 2013 she founded MediaWallah based on this conviction. Amidst growing challenges of walled gardens, data fragmentation, and “black box” data providers, MediaWallah would develop groundbreaking identity-by-design solutions as the answer. Nancy has played a pivotal role in shaping some of the most crucial tech players in the marketing field. Prior to MediaWallah Nancy was CRO at tag management company TagMan (acquired by Ensighten) overseeing sales, marketing, product implementation, client management, and business development. Prior to TagMan, her positions included VP of US Sales at IgnitionOne (acquired by Dentsu), VP of Sales at x+1 (acquired by
Rocket Fuel), Senior Director of Strategic Development and Partnerships at DRIVEpm (acquired by Microsoft) and Senior Director of Ask Jeeves (acquired by IAC/Interactive Corp).

Previous ArticleNext Article