Trading Desks are in for the long haul, not the sale.

I cant decide where to start on this post, there has been so much going on in the hectic world of ad exchanges in the last few weeks. Top of the bill was an excitable debate between an Audience on Demand employee and a disgruntled DSP. The key issues raised around conflict of interest included agencies being forced to put spend through their trading desks, lack of impartiality etc etc.

Interwoven with this debate was the fact that so many companies are approaching us at the moment, DSPs, Data targeting companies etc all with interesting premises I suppose but all with one thing in common, they all need to make as much money as possible, as fast as possible. Lets talk about conflict of interest..I use the DSP marketplace including Triggit which was involved in the above debate. How many shall we say there are, that are currently aiming for Trading Desk revenues – 4? 5?. Everyone is coming to town, everyone wants a piece of the action, but when they get into town they realise that a couple of those 4/5 have been busy for a few months / years and pretty much wrapped up the business. Its not to say thatagency groups will not test and learn, we do in the US and there is definatley room for more than one or two but for some, the market’s not big enough. What happens then? They need to fight for revenues, they need to say why they are better than each other and especially better than Invite to try and find the big ticket, except I am not sure there is a big ticket at the moment. So then they resort to the last option which is to try and undermine the credibility of a trading desk to try and open up some cracks of opportunity.

The conflict of interest for those guys is they have to make money to keep the VCs happy. The agency group trading desk model is not in the same boat. Audience on Demand’s sole purpose in life is to navigate on behalf of its clients a very complex market place and deliver great results. They are in it for the long haul, they have much more to lose. AOD messes up on a client it can jeopardise the whole business. Yes there is pressure to deliver..but its to deliver results not revenue first and foremost. In a competitive marketplace as the agency landscape is, the more things you do well and right, the more chance you have of retaining the client.

So whats better then? An organisation like Audience on Demand that has a remit to make sure it is working with the best, understanding strengths and weaknesses – and believe me all these tech companies have them – or a heavily invested tech company struggling to make ends meet. Who is actually going to have the interests of the client? I can tell you, it’s us. Anyone who thinks that agencies and clients are naive enough to accept sub standard strategy and results just because its in house is a) clearly lacking in understanding of how an agency works and b) underestimating the clients and Account people. If a client asks about our impartiality we can show them the full vetting we do of all DSPs, I can show them the data compliance methods we have in detail for every supplier, I can show them the results in detail where an acceptable flat cpa or cpc is not acceptable as it encourages the supplier to focus on growing their margin rather than delivering the lowest metric. I will show you 100’s of people who live and breath this space and understand it better than any individual tech company thats trying to undermine it.

Conflict of interest is doing what you have to do to stay afloat in one of the most competitive eras of all digital times vs doing what’s best for our clients. Finally it is always worth analysing who is throwing the mud, its often one of those people who came in to town too late and cant find anywhere to hang their hat.

10 thoughts on “Trading Desks are in for the long haul, not the sale.

  1. And that pretty much sums it up. Whether a trading desk is right for you is a question to be answered between a client and their agency. Anyone else’s perspective/speculation/concern is irrelevant and depending on the source, conflicted heavily be self interests themselves. As the aforementioned company is. I’d love nothing more than to see my operation be in partnership with the very past technology companies in the business. So far we are. For those who have work to do to become best in class, at par or better than the leading DSPs I suggest they put their head down and get after it versus what we’ve scene from several to date – lot’s of smoke, zero fire.

    Great post Marco

  2. Pingback: U.S. Justice Department Still Eyes Google-Admeld; Interclick’s Ad Genome Focus On Data

  3. Pingback: News of the Day | | online advertising news, trends & industry talk

  4. Hi Marco,
    Thanks for posting this. I found this a very valuable read. Instead of just agreeing or disagreeing, I thought I put some thoughts into my reply.
    As you know we have discussed several times about our different business models, lately at the NMA Event. But, reading your post, we are actually fighting the same corner.
    As a trading desk of course, you would ask your media buyers to put spend through your trading desk rather than going through each individual exchange and DSP buy, or an external trading desk/DEM. I guess there are two reasons for that, one being the revenue and two being the control.
    Looking at the landscape, more and more DSPs entering the market, and given the outlook of ad servers out there and with them developing into DSPs and SSPs, we will see more technologies entering the market over the next 12 months. Each of them will want to have an equal share, each will say they have the best technology and algorithm/decision engine.
    Unfortunately, it still comes down to testing to decide which technology is or isn’t worthwhile using, and which one adds extra value. DEM specialists like mexad, are testing all DSPs out there, offering advice to agencies and integration with trading desks.
    Too often trading desks are tied in or want to be tied in to one technology. By working with a service provider (DEM) in close alignment, you are able to get learning from other DSPs, see which ones performs best for your clients, and then decide which one to use and which one not to use.
    Your US office from what you say are exactly doing what mexad does, testing all technologies. And, you say they are advanced, hence I take it as a compliment 😉
    I will publish an article next week in iMedia stating, that relying on one DSP just doesn’t cut the mustard. The only way currently to get the best performance in the ad exchange and RTB market is to use several DSPs and technologies alongside each other, utilising their different bidding algorithms, learning, decision engines and RTB capabilities in order to get the best performance for any advertiser’s campaign.
    I know the biggest fear, coming from agency side myself, is to outsource RTB buys to specialist, then buying back the expertise you lost – similar to search. But display is different. There are less limitations, and we can all happily work together and alongside for the greater good of advertisers, creating better value for your clients. This wouldn’t be possible in search.
    Of course what I would be cautious about are the “networks using DSPs for additional reach”, or DSP service providers adding publishers, offering an outsourced media buy. They are competitors for you. We are just like you, fighting the same corner.
    Looking forward catching up with you again soon. Thanks for the article!
    Have a great weekend.
    Best, Volker

  5. In a perfect world, Mac and Marco’s arguments would hold true. I don’t work at an agency but have been in the industry a long time and I have plenty of close friends on agency media teams. I get all the feedback directly from those in the trenches. I often hear of planners being pressured by heads of agency’s to put their trading desks on plans, despite lackluster past performance. We’ve all been around the block a few times and it smells an aweful lot like kickbacks and large bonuses are involved on cashflow that goes to the trading desk. You can have the smartest people in media but it’s human nature to be tempted by huge sums of money for opportunistic gain. And given that the agency controls the media budgets, there is natural tendency for favoring their own entity and thus a conflict of interest.

    • Thanks for the post media gangsta, I can only say that its is not correct, our performance determines everything. AOD has dropped off a number of campaigns where we have not delivered, there is no way for me to prove it, but rest assured I have been through the mill to get AOD on plans. Nothing more I can say on that. Thanks for the post.

  6. This topic is very interesting, I can say that as a client I’d rather have an agency that does not make money by recommending this or that. I have seen situations where the agency is pushing to include their trading desk on the plan and not even considering testing one or more options and see what performs better. That in my book is the result of a conflic of interest. As a client I am savy enough to push back and say hey why don’t we try this other thing too, but why should I be the only one interested in see what perform best, isn’t it the agency’s role too? At least there should be clear policies on how to ensure the conflic of interest does not affect your clients. How do you ensure that in your company? I would love to hear and see if there is something we can do since obviously agencies are not going to get out of the business of trading ads.

    • Thanks for the post, great to have client feedback. One point that is missed in this debate is that we work with a number of DSPs so we do indeed see what performs and what does not. We also have Vivaki Verified which vets the DSPs, Innventory partners and data partners. Therefore to your point about agencies should test a number of companies, that is exactly what we do, it has been lost am afraid. Thanks for the post

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s