What Is A Mobile DSP?

mobile-DSPSome companies claiming to be “mobile demand-side platforms” are veiled mobile ad networks, or at least they’re mobile ad network-like. Which is to say they may be taking margin without disclosing price paid to the client.

Others may be more transparent, buying on a per-impression basis and charging licensing fees as a standard DSP would do. These companies may bring bidding and optimization features, and data sources, geared to capturing impression opportunities among the confusing array of smartphones and tablets.

We asked a few people for their take on this frequently applied label.

Jennifer Lum, Co-Founder, Adelphic Media

“Mobile DSPs need to unlock the marketing potential of the largely anonymous mobile inventory available in market today. Mobile DSPs require advanced technology that enables faster and more complex processing on a much larger data set. Mobile DSPs need to collect, combine, analyze and process massive amounts of real-time data in order to execute successful campaigns.  Decisions must be made based on real-time analysis of over 100 mobile data points.

There are factors unique to mobile that DSPs must be able to handle across operating systems, browsers, apps, locations, and thousands of devices. An effective mobile DSP must offer robust solutions for each of these fundamental challenges:

    • Variability in data availability
    • User identification
    • Audience intelligence and segmentation
    • Tracking

Of course bidding, targeting, and optimization tools are also required. However they must be designed to consider mobile-specific data to drive real-time consumer behavior and positive ROI.”

Paul Dolan, Managing Director, North America at Xaxis

“Mobile demand-side platforms essentially should not exist as a standalone entity. Rather, the buying and optimizing of mobile media should be a part of any robust DSP.

Mobile does present several unique opportunities and challenges that need to be addressed, including:

  1. Inventory. In the absence of a privacy-compliant identifier, much inventory is still siloed in mobile networks.
  2. Creative ad units. Size, vendors and device capabilities need to be considered.
  3. Tracking. Whether it’s an IFA in iOS, a cookie in a non-Safari browser or some other persistent identifier, frequency capping needs to be used.
  4. Mis-taps. Click/Tap rates tend to be sky-high in mobile, especially in-app, but much of this tends to be mis-taps and not legitimate traffic.
  5. Targeting. Location can be a powerful tool and is carefully used.

Much like Facebook’s Ad Exchange, just because something is different doesn’t always warrant a separate platform for programmatic buying. All of the above items can easily be built into existing solutions. Otherwise, you would have disjointed media planning and execution resulting in little opportunities for scale.”

Elizabeth Zalman, CEO, Media Armor

“I like to think of a demand-side platform as one place in which a marketer or agency can implement and execute all display advertising initiatives.  What is a mobile DSP, then?  If we follow the definition above, it would be a place to go to source mobile display.  Two questions then arise: Mobile?  And why?

What is mobile?  It’s not a channel, and it’s certainly not a device.  The channel should instead be defined as digital, and the devices within digital are desktops (including laptops), tablets, and smartphones.  Our industry is confusing devices with channels.  If mobile isn’t a device, and it isn’t a channel, then how can we create a specialized technology which enables us to source display for it?

This leads us back to the question of ‘Why?’  I believe we would agree that there is an incredible amount of fragmentation in digital marketing initiatives.  Given that, why would a marketer want a place where they can only buy one type of (mislabeled) display?  Ultimately, if pure mobile DSPs do exist, they do not solve for the reason that willed them into existence.  They instead exacerbate a pain point for the demand side.”

Enjoying this content?

Sign up to be an AdExchanger Member today and get unlimited access to articles like this, plus proprietary data and research, conference discounts, on-demand access to event content, and more!

Join Today!


  1. Two of the three contributors say mobile should not be standalone, but a component of digital. In an ideal world, yes – but where online DSPs lack the functionality fully to exploit mobile as a route to reaching consumers, mobile DSPs provide a powerful option. As the market matures, there is bound to be consolidation between the desktop and mobile spheres, and online DSPs will have to decide if they should build, buy or partner to strengthen their products – but that takes time. The same is happening in the agency world with the mobile specialists (and happened 15 years ago when traditional agencies couldn’t service online).

  2. At RUN, we definitely agree with a few of the points raised here and some of the attempts to define the “Mobile DSP,” especially around consolidation with one partner. In fact, we built our product suite with this ‘one stop shop’ mentality in mind. But in reality, there are no horizontal DSPs that solve everything across all verticals, optimally. This is definitely true with emerging channels, devices, and sessions, especially for mobile and the fact is it requires/deserves a specific focus to do it properly. We have built up plenty of case studies and insights through our direct hand’s on platform experience that would support this.

    I find it overly presumptive—if not borderline uneducated—to say that all DSPs are simply a place to “source display.” That definition may apply to some legacy systems, but I can assure you from my experience that innovation in this part of the ecosystem does exist, and that the next wave of programmatic solutions are being designed with that exact perspective in mind.

  3. Brett Whitehouse

    I would argue that consolidation is happening sooner than you think. The major web DSPs / SSP’s are already forging links with the major sources of mobile inventory. Mobile is just another UA and I would be extremely nervous if I was a purely mobile DSP.

  4. A Mobile DSP is exactly what you should think it is – a platform to buy and optimize mobile RTB inventory. And have a single focus to innovate on mobile-specific problems that arise in this wild west of mobile RTB.

    As Seth and Jennifer mentioned, just some of the problems you must address are conversion tracking, creative unit variations, audience and segmentation data (or lack thereof!) and consolidation and standardization across exchanges/SSPs.

    It’s absolutely offensive to hear and think that we are just slapping a “label” on our platform and technology. Ad networks are a dying breed unless they innovate into the RTB space – and we have seen them pick up DSPs recently with Millennial acquiring Metaresolver.

    And as such, at PocketMath we carry no relationship with Publishers (apps, etc) and therefore have no conflict of interest such as an ad network would. Our focus it to give advertisers a place to buy mobile RTB efficiently through one single source. And get full transparency on every ad placement served. Innovation is happening.

    • Eric Steinhardt


      What is coming across as “absolutely offensive”? You openly state that a Mobile DSP is a platform to buy and optimize RTB inventory. What is RTB? It’s an exchange of data server-to-server, and being “mobile” doesn’t make that exchange of data any more unique. Instead, it saturates the industry by creating additional silos, under the guise of differentiation, where marketers are looking to source “digital” display. Would you prefer to buy your milk and cereal at different stores?

  5. With global Internet users doubling in size over the next three years and most being mobile, the “channel” (or whatever we want to label how we are accessing and using the online universe) will represent the lion’s share – data subscriptions will outnumber fixed Internet connections by at least 4x by 2015. Anyone in ad tech will agree that the data and features of mobile are vastly different than display. Therefore, the tech stack needs to be specific to mobile. At least when done correctly. And when done correctly, innovation can happen much more rapidly than if you are bandaging/retrofitting legacy technology for desktop advertising By doing so, advertisers will advantageously correct the massive gap between time spend and ad spend. At some point the “channels” may converge, but only when technology gets us there. At Gradient X, we believe it takes a mobile centric approach to unlock the promise of mobile today.