Login
Login is restricted to DCN Publisher Members. If you are a DCN Member and don't have an account, register here.

Digital Content Next

Menu

InContext / An inside look at the business of digital content

Fact check series, part three: What really defines a next-generation DMP

July 6, 2020 | By Adam Solomon, CMO – Lotame @adam_solo

DMP 1.0: Contextual-to-audience targeting for desktop web & DFP

Since the advent of newspapers in the 17th century, digital advertising has been premised on developing and promoting content to attract audiences, and then selling contextual ads against such content. Even the arrival of online services like AOL and the disruptive nature of the web didn’t materially change the contextual nature of advertising against publisher content.

Approximately 15 years ago, digital publishing experienced a seismic shift. The capabilities afforded by web browser technologies and advancements in server-side data processing introduced audience targeting. Ad networks and technology platforms could create rules to associate consumer web surfing activity on publisher sites with behavioral attributes. These could be attached to browser cookies. And they could target those cookies with advertising on websites other than the ones where the original behaviors were observed.

Data Management Platforms (DMPs) arrived to help publishers save and organize all the data created by digital traffic and to take advantage of audience targeting.

The role of the DMP was pretty straightforward: Transform contextual desktop web activity into audience segments and pass those audience segments into a website’s adserver, most likely DoubleClick for Publishers (DFP). Life was simple in DMP 1.0.

DMP 2.0: Mobile and social and programmatic

Five years later, smartphones, social networks, and programmatic media buying shuffled the deck. A new breed of DMP company arose that specialized in audience targeting for mobile app environments. Social networks introduced their own audience targeting. And programmatic advertising ushered in the need for data pipes connected to buy-side platforms.

An important challenge was how to use data to effectively plan, activate, and analyze media campaigns across channels and platforms. This led to a focus on the mechanics of syncing IDs between platforms using in-page pixel syncs. It also led to the emergence of technologies and venture-backed companies focused on “cross device” or “device graph” technologies. The primary challenge was how to associate activity on a desktop/laptop computer with a mobile device – literally “cross device.”

Under the hood, the reality was more nuanced. When it came to desktop/laptop computers, cross-device technologies were using 3rd-party cookies in web browsers as profile identifiers. So, if a person was using two different web browsers on the same computer, it could conceivably be considered two devices. If a person deleted their 3rd-party cookies, and a new profile identifier was generated, then it would also be considered a new device.

On mobile, there were other important nuances regarding what was considered a device. Mobile web browsing was similar to desktop web browsing in terms of its reliance on cookie-based identifiers. In the mobile app environment, both Apple and Google moved towards providing OS-based solutions for provisioning consistent ad identifiers to all licensed apps on the device. Apple iOS called this IDFA. Google dubbed it GAID. It’s significant that the hardware guys, who in some cases have declared war on cookies for mobile web, have facilitated a static advertising ID for mobile.

So “cross-device” was — and is — more like “cross-ID” based on the underlying reality.

In addition, cross-device technologies introduced new vocabulary. Deterministic and probabilistic matching described how these IDs could be associated. These syncing and graphing technologies were nascent and not well integrated across the ecosystem. However, they did spotlight the need for publishers, marketers, and ecosystem participants to look beyond desktop web + DFP (DMP 1.0), and to either build, buy, or license technologies to connect data across devices and media buying platforms.

DMP 2.0 was defined by data management solutions that began to address the emergence of additional device types and media buying platforms. It was also characterized by the technologies required to connect and extend audience segments across those environments. Even back then, publishers knew that consumer data needed to be fluid and connected across platforms and partners in order to drive business forward.

DMP 3.0: The DMP is dead. long live the DMP!

Media and technology have continued to change rapidly.

On the data input side, publishers are seeing consumer 1st-party data flow into their systems from web, mobile app, email, CTV/OTT, CRM, and an assortment of smart/IoT devices. They also have the ability to access consumer behavioral and attribute data from 2nd-party and 3rd-party sources. On the output/activation side, publishers feed data into adservers, SSPs, programmatic header wrappers, analytics platforms, content optimization platforms, content management systems, data lakes/streams/rivers, and marketer platforms.

an exploding array of consumer devices and applications, are powering these data flows. Unfortunately, most were not designed to be interoperable from a data perspective. To add further complexity, browser developers such as Apple and Mozilla block (or shorten the lifespans of) different flavors of browser cookies to better safeguard consumer privacy. As a result, they make it more challenging to connect data in web environments. Hanging over these operational data challenges are regional data regulations that provide consumers with protections/controls over how their data is tracked and used.

Innovation and M&A and corporate development activity have grown in the data management space as companies look to expand their solutions. Some pure DMPs joined forces with larger martech companies and integrated into their stacks. Others maintained independence and focused on developing advanced data-driven solutions to specific challenges.

Against this backdrop, it has been increasingly difficult to define what a DMP is. Today, client types, data types, and use cases are incredibly diverse. A single acronym isn’t sufficient to capture the full scope of capabilities.

But whether a company describes itself or its solutions as part of a DMP, CDP, martech stack or other, it’s clear that effective next-generation data management technologies must provide publishers with scaled and privacy-enabled connectivity between data sources and platforms.

Who wants to live on Gilligan’s Island?

It’s surprising to see that the recent crop of DMP start-ups have developed technologies and solution strategies that focus on publisher isolation instead of publisher-to-ecosystem connectivity. They have essentially stranded publishers on their own “data islands” in structures that might not survive the next privacy or browser storm.

They encourage publishers to turn the clock back to DMP 1.0, transform contextual desktop web activity into audience segments, and pass those audience segments into a website’s adserver. These contextual DMPs don’t have any magical technologies. And the functionality that they provide are features (or sub-features) of much more complete and robust data and marketing technology platforms.

These “disconnected DMPs” completely lack a focus on the capabilities required to help publishers and their marketing partners stitch data across websites, devices, platforms, and channels. It is imperative that any modern data technology platform provides scaled data connectivity solutions. DMP start-ups lack ID graphing technologies. To the extent that they have any inter-platform connectivity, they have to rely on 3rd-party pixel syncs powered by other companies. (Something they don’t talk about publicly since it runs counter to their cookieless fiction.)

To find your people, connections are key

The challenges publishers and marketers face in a fragmented data landscape are significant … but not insurmountable. Isolating your data from the broader marketing ecosystem is not the right path forward for publishers looking to grow their businesses.

Complete, consistent, and compliant data connections will deliver greater scale and precision for brands and publishers and better experiences for consumers. Publishers need next-generation DMP solutions made for today’s digital advertising challenges and future-proofed for tomorrow’s opportunities. They need trusted solutions grounded in partnership, privacy, and a panoramic view of the consumer.


About the author

As Chief Growth Officer, Adam leads Lotame’s global marketing, product, and client success teams to support growth across customer acquisition, retention, and revenue. He joined Lotame as the Chief Marketing Officer. His diverse experience balances art and science, and includes stints as an aerospace engineer and patent attorney, plus 21 years in consumer media and marketing technology in leadership roles at Viacom Media Networks, Time Inc., Hearst, and PebblePost. Adam is co-inventor on four issued U.S. patents related to interactive video advertising technology.


Fact check series, part two: contextual targeting, audience targeting, and match rates

Fact check series, part one: the truth about “cookieless” DMPs

Print Friendly and PDF

Liked this article?

Subscribe to the InContext newsletter to get insights like this delivered to your inbox every week.