Ad Agencies, Exchanges And The Disintermediation Opportunity

A flood of behavioral targeting company news coincided with the OMMA Behavioral conference at the Marriott Marquis in NYC this past Thursday.

In ClickZ
, Michael Katz of interClick noted the recent appearance of new data “intermediaries” – such as yield optimizers Rubicon Project, “meta network” MediaMath and ad data exchanges like Media 6° and BlueKai (who also announced a milestone in its eCommerce data for online ads).

Katz offered:

“There are many more intermediaries becoming involved in the supply chain. I don’t know if it’s a bad thing, but rather born out of necessity. It will be interesting over the next year to see at what point as you continue to add intermediaries will disintermediation start to occur.”

It’s funny you should mention disintermediation, Michael. Some think it’s already rolling.

During the OMMA event, Darren Herman of Varick Media Management and Media Kitchen fame twittered “ed montes (Regional Manager, Havas) just pissed off every ad network in the room.” And then added, “I totally agree with him as he said that ad agencies are acting as tech players now and disintermediating.”

From our perspective, distermediation makes great sense for the media agency model. But, we don’t believe many of today’s agencies are going to be able to pull it off with their anachronistic, walled garden mentality.

The growing number of tools of the exchange for advertisers, publishers and, ultimately, traders, creates an opportunity for an agency to provide expertise across the exchange’s range of tools which cannot only affect efficient, ROI-generating strategies for clients but also for their own book. All new revs for the savvy agency!

The trip wire for certain agencies remains a culture of entitlement and a self-proclaimed need to protect data on behalf of their clients. The inevitable success of the ad exchange model will be based on the free flow of data between all sides of the exchange as opposed to the type of false disintermediation which may tend toward proprietary data grabs.

Social Media – A Forerunner of Ad Exchanges

Much like social media’s success in empowering the individual to create his or her own content with the people, places and brands that only matter to THAT individual, the ad exchange model will flourish as disparate parts of data can be freely coalesced by any exchange member and provide transparent insight into the true value of media.

This “true value” will result in vast new quantities of premium inventory and bring the exchange out of the perceived shadows of the remnant world and into the rarified air which approaches every marketers wet dream of one-to-one marketing as buyer and seller know what they are giving and getting – and the consumer receives more relevant advertising.

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. The big media buying agencies will be reluctant to throw their lot in with the ad exchanges.

    Their whole model is based on media bulk buying and media spend commission. They would have to restructure their entire business.

    It’s probably why you are seeing the early succss of agencies like Varick and Infectious.

    I’m interested in the last two paragraphs, Joe:

    “…the ad exchange model will flourish as disparate parts of data can be freely coalesced by any exchange member and provide transparent insight into the true value of media.”

    Are you saying that the market will ultimatley establish the true value of advertising on the exchange? And do you think if exchanges operate independently and publishers put inventory on multiple platforms, there might be opportunity for arbritrage?

    Will you then see the rise of vertical exchanges to combat abusive arbitraging? Nasdaq lists tech stocks. Could we see a pure tech ad exchange? Or is this just another intermediary in the supply chain?

    • Thanks, Ciaran. First, the exchange isn’t an intermediary. Quite the opposite in my view. It’s agnostic and generates revenue through transactions and/or membership fees… As for your other questions, let’s get out the crystal ball!

      Yes – tools which the exchange offers will provide insight on true value in the marketplace – ideally with insight variables on past, present and predictive, future demos, geos, context, performance and anything else . As always, for any advertiser campaign, the wild card will be the advertiser’s creative – Randall Rothenberg can rest easy about the continuing need for great creative. Technology is the creative’s best friend. A/B testing will become be A-Z testing.

      There will absolutely be arbitrage opportunities inter- and intra-exchange. I don’t see any reason there wouldn’t be – its not a question of abuse, its a question of the free market and providing value. I’m curious how you see vertical exchanges developing. I can see vertical expertise on the large exchanges where a contextual technology overlay is used rather than any vertical exchange per se.

      As for units of inventory bought and sold like Nasdaq listings – why wouldn’t there be a “listing”, so to speak, for an ad placement, behavioral data, and so on.. I just think it won’t be something you’ll look for in your Sunday paper and you’re going to “own” blocks of impressions not part of a company.. there will be billions, if not trillions of so-called “listings” that through technology are assigned value by buyers and sellers in the marketplace. Open APIs into these “listings” will be available to all exchange members who will undoubtedly be using their own proprietary system or that of a service provider in order to buy and sell.

      • Thanks Joe. I appreciate you addressing my questions. I suppose when I was alluding to new exchange verticals, I was probably thinking of industry verticals, which on reflection would limit the model’s effectiveness.

        I can’t help thinking there a lot of exchanges out there at the minute, and that only a handful will be left standing. But these are interesting times for the industry.

        We are a little bit behind the curve in the UK, but I’ve seen a couple of interesting companies here trading on the ad exchanges.

        I would love the opportunity to work with a company like Varick, or any other company trading on these new platforms. I am available at any stage to discuss this. Cheers Joe.

  2. During the OMMA event, Darren Herman of Varick Media Management and Media Kitchen fame twittered “ed montes (Regional Manager, Havas) just pissed off every ad network in the room.” And then added, “I totally agree with him as he said that ad agencies are acting as tech players now and disintermediating.”

    In response to this, it will not be that simple for agencies to disintermediate ad networks anytime soon. There is an inherent DNA issue that prevents agencies from going around networks anytime soon (not to mention a fundamental technology gap). The agency model is a service model while the ad network model a supply chain management model. Agencies simply do not have the tools or the expertise to execute a supply chain play. I would be happy to share my thoughts even further on this but I would predict that large holding companies will start to align themselves and ultimately acquire ad networks before they start to circumvent them. Furthermore, look for networks to start disintermediating agencies. Look at Specific Media’s relationship with Ford over the past couple years. This was not an anomaly but rather the beginning of a paradigm shift.

  3. There are going to be very few agencies who can pull this off at scale. Afterall, agencies don’t typically build their own technology, and some of the first who do will fail at it (not all, but many). So if there aren’t enough acquisition to go around, then the time that it would take agencies to disintermediate will be long.

    Building Virtual Brand Networks or something similar for clients and being able to action them across exchanges and other media sources is going to be very important to align brand goals.

    After all, show me ANY ad network that is fully aligned with a media agency and client.

  4. It is not hard to imagine agencies starting to build their own mass reach behavioral networks – look at DrivePM. Many networks want to believe that by aggregating mass reach they are creating value where the trend seems to say otherwise.

    It is too difficult to talk about ‘ad networks’ as if they are all one large blob. Companies with specific optimization technologies will continue to succeed in specific categories – TURN, Quinn Street, etc…

    Companies with great reach and deep integration across large numbers of websites will continue to succeed in specific verticals – Glam, GayAdNetwork, FM, MomLogic, etc…

    Agencies can replace mass reach vehicles for retargeting – they have better access to create segments with their clients data and better negotiating power, especially in this market.


  5. I’m not sure why most agencies would want to fully disintermediate, since that would mean taking on all of the network-growing responsibilities of a network or exchange.

    The real value that a network or exchange provides to the end-advertiser or the agency is that they’ve already done the hard work of building scale on both the publisher and advertiser side. By bringing both ad supply and ad demand to the table, the exchange / network allows their advertising clients to focus on adding value by testing different targeting options, messaging, creatives, etc. Those are the functions where an agency can add tremendous value, because presumably they’re directly involved with developing their client’s brand identity in the first place.

    The other point I would add is that there are now so many exchanges and networks that any advertiser or agency can try out for themselves, and start using within minutes. The role that an agency can play in helping a client navigate these choices is similar to the role they’ve historically played in TV placement. Agencies help manage the creative aspects, and they also help manage media planning across platforms so that the client sees the best ROI possible.

  6. I tend to agree with Michael Katz. I am not so sure that agencies in general will be able to disintermediate ad networks with the help of exchanges. Their DNA is client service NOT technology. Rather the differences between ad networks and agencies will become fuzzy. I can see smaller agencies becoming media brokerages and bigger agency groups acquiring ad networks and turning them into trading platforms who buy of several exchanges. I also see bigger ad networks becoming transparent and building out client service divisions resembling a media agency.

    It’ll stay interesting for some time to come.

  7. The big agency groups will take time to re- engineer their infrastructure, process and skillsets around a more data and tech driven approach. They are simply too large and tied into legacy models to move quickly. The stuff we’re talking about here requires such a ‘root and branch’ upheaval of the traditional agency model of ‘Bigger buys cheaper’. As we all know in an auction, ‘SMARTER buys better’. For this reason, i think it will be a new breed of smaller smarter tech savvy agencies that will lead in this space in the next couple of years, and they will have a strong enough proposition to win DR/ROI focussed budgets from th big guys. Hopefully Infectious Media will be one of them.

  8. I think that one cannot underestimate the power media agencies have through their client relationship. Therefore I think there is definitely an opportunity for them to disintermediate with the help of exchanges. I just think that they will use existing tools and / or acquire companies with technology in most cases rather than building it inhouse simply to be faster.

    • To my mind the whole situation sounds like 2001-2002. Google came in to play with AdWords & small SEM´s started to roll up the field from behind. The client relationships didnt count that much and the big agencies lost most of their clients search budget to SEM´s. They all grew like hell until 2006/7 – when they where bought by the big agencies. I dont think that the big ones will be able to re-engineer their infrastructure. They will try (and they currently try hard) – like 2003/2004 – will fail, wait & buy in around 3-5 yrs the succesfull, new technology driven agencies.