The ‘In-House’ Trend: Can It Last?

joannaoconnelrevised“Marketer’s Note” is a regular column informing marketers about the rapidly evolving, digital marketing technology ecosystem. This week it is written by Joanna O’Connell, Director of Research, AdExchanger Research.  

In a recent interview with CMO Today, Sir Martin Sorrell, CEO of holding company WPP Group, called the “in house” trend (i.e. marketers moving media buying – driven by the rise programmatic – out of agencies’ hands and inside the walls of marketing organizations), “a temporary phenomenon”, noting, “Our view is after a year or two it will change. I question whether [clients] will be able to apply technology successfully.”

It makes me think of two questions I raised at the end of the recently released research report on this subject, which I share here because they are apropos to Sir Martin’s argument:

What is one’s definition of effective? There is general agreement among marketers choosing the “In-House” path that they can do great work themselves, but a counter-argument that any programmatic shop (or holding company trading desk) would make, which is, frankly, compelling, is, “Can a small, lean group of marketers who are not steeped in programmatic really do better than what we could if given the chance?” Is there a performance trade-off to having one supply manager negotiating rates versus a whole team, or one platform expert versus a whole host of experts, regardless of money saved through fees not paid out? Asks one director of advertising and online marketing at a leading catalog and e-commerce retailer whose organization is currently migrating to a more complete in-house model, “How do you know for sure you can do it better? You don’t. And the arrogance of thinking you can do it better can be risky.”

Must the definition of “agency” necessarily evolve? Also at play is the evolving definition of “agency,” where the lines between the classic media agency and the new breed of service partner, such as from an independent programmatic shop, or even a DSP with a managed-service arm, continues to blur. It’s clear that even the most well staffed in-house media teams rely on external partners, whether for executional or strategic help. We live in an era of holding company trading desks, where debate continues to rage as to whether the model itself, or some of its specific practices, such as taking a principal role in media, are inherently at odds with the traditional definition of what it means to be an “agency” – and while the specific model of centralized service-based desk itself may not have long-term legs – the questions its advent raised just might.

Long and short, these may be philosophical arguments at their core –

  • Does a marketing organization value 1 percentage point better performance if it doesn’t trust who’s managing its media dollars?
  • What does a marketing organization believe an agency is really there to do for it, and is that marketing organization open to evolving its thinking in the face of such significant change?

This brings me back to Sir Martin’s position on the future of “in house” media management.

It’s a fascinating question: whether or not marketers will “cave” and return to agencies (because that’s basically what he’s implying). He might be right. He might not. He’s certainly right that client-driven margin pressure has been hard on the agencies* – it’s no surprise the holding companies doubled down on the higher margin business that is the trading desks; they are seeking financial success, as any good business does.

But it’s an awfully big gamble to take. Because adherence to the holding company party line without acknowledging – and working to come to a mutually agreeable solution to – the problem they helped create for their agencies and clients could prove a dangerous game.

As always, thoughts and comments are welcome.


*It’s funny, I wrote a whole Marketers Note on this just a few weeks back and it seemed to fall flat. At the time I wondered – is it that the subject is too hot to touch, or are people so bored with it that they don’t bother engaging? Likely, it’s neither – it’s that it’s a problem without any easy or obvious solution.

Follow Joanna O’Connell (@joannaoconnell ) and AdExchanger (@adexchanger) on Twitter. 

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. I think the core issue here is trust.
    Advertisers want agencies to show them “debundled” costs of the ATD business, because they believe they have the right to know. They’re also concerned whether they’ll paying agency fees twice (one for a media agency and the other one for an ATD). Agencies (actually holding companies) refuse to show them these costs, because they believe that will make them vulnerable and will result, as they saw it with the traditional media, in advertisers demanding agencies to cut their fees or maybe even remove them all together (“we are already paying for the media agency, why should we pay more for the ATD services if you’re a part of one holding company”). Advertisers can take programmatic buying in-house and there is no question that a certain result can be seen, but I also believe the costs will be higher than they think. They need to invest in the programmatic buying infrastructure (not a small investment) and I do not think they can get the same prices as ATDs, since the latter have bigger volume.

    On the other side, agencies are under huge pressure to cut their fees. For some reason, many advertisers do not think they should pay decent fees for good media services that will result in meaningful results for their brands. Many advertisers may very low fees and postpone payments for a very long time, which means that the operations basically have to be funded from the agency’s pockets.

    There is no easy solution.

  2. happy the one who knows the end of the story..

    now..i tend to agree with M. Sorrell..and it’s my opinion since years..
    – many time i’ve met with people at clients willing to do it in house, we were facing people who believe they are superior in every aspect to the media agency people..
    they believe that with their unique brand name and budget they’ll be able to put some pressure on the googles or facebook..and have better pricing than the agency (which represents 100 or 1000 of clients..) the math….at least i know google does not care at all..neither turn by the way..
    – other points is that this market is really complex to manage..i tend to ask them one unique question : which DSP and why ? which data provider and why ? do you have people in house to do that, people who will be better than people doing this since years with historical data and statistical reasons to prefer this one rather than the other..
    – other point being that programmatic is mature for ain’t for TV, radio, DOOH, print do you intedn to do inhouse only the display side of your business or do you intend to do everything..just asking because you know these are not the same jobs..and by the agency will (again) certainly have better rates than you will by yourself with TV or radios or magazines..
    – and other point is, that the clients hasn’t figured out how to track cross device..(oh maybe he think he has..because he believes the speech of all those vendors..) just because it’s not (yet) working.. so happy the client who knows how to track business accross all medias by himslef
    – By the way Mr clients willing to do it in house..have you figured out, that in fact doing global programmatic capaigns is not just about pushing a button or uploading your creatives and choosing a cpm..did you notive, that latam is not as mature as EU or USA ? do you know that china is a walled garden fact do you know that in fact you’ll need tens or hundreds of local people to do that ?

    finally, i’d say that there will be quite a lot of disapointment (and people fired bythe way) just because they wanna test it..what for ??

    they’d better think about the way they are paying their agency..
    or maybe look at how well the google/facebook/criteo are going (financially speaking) and compare that with the financial situation of their agency..

    i believe, the clients have created themselves this situation where they are thinking the agency’s robbingthem, whereas they do not question the pure players who are doing 60+% margins publicly..

  3. Joanna – nice piece.

    Sorrell’s perspective fundamentally misses the point as to why these marketers are taking this function in house, based on our own discussions with marketers.

    They are not taking the programmatic infrastructure in house to save 1% on costs, or because they believe their people are inherently smarter than their agency peers. It is because:

    – First and foremost, they want direct control and access to the data generated by these campaigns, with the belief that this data will deepen their understanding of their customers – the CMO’s #1 objective in pretty much every CMO survey. Even if they know they can’t do any better today than their agency in extracting value, they know that if they own the data they can use it tomorrow when they are smarter.

    – Chickens have come home to roost on an underlying trust issue. There’s a close correlation between this increased interest in moving platform ownership direct, and the strong squeeze that agencies have underway to force all media buying through their trading desks. While there has for some time been an underlying concern that the agency may not have interests fully aligned with the client, this squeeze has exacerbated that concern.

    While i don’t believe the marketing organization thinks it’s smarter than the agency, they also don’t believe they’re somehow intrinsically dumber… marketers are making a calculated bet, i think rightfully, that over time they have as much chance as an agency of hiring the analytical/data savvy employees they’ll need to extract key insights from their execution activities.

  4. The question may not be whether Sir Martin is right or wrong, it may have to do more with what type of industry have we sown that after billions of dollars of investment, what we are left with is an environment of such mistrust by the two participants that matter most – the advertiser and the publisher – to cause them to want to go it alone?

    After years of innovation we are at a point where the people with the money, the advertiser is separated by billions of dollars of technology from the people with the audiences, the publishers, ironically and magically achieved with the promise of bringing the two closer together – “at scale”.

    This billion dollar separation is cloaked in jargon and gibberish to help with the confusion that only the middle men with cloaked methods can unravel. Marketers and publishers are waking up to the fact that they are getting fleeced so the magnetic pull to control their individual destinies is driven by a realization that much of what is being said is in fact false and detrimental to their business model.

    So the question is, what can we do, as an industry to create a culture of trust so the need for an explosion of singular efforts is stemmed? Additionally, in an industry that thrives on innovation and disruption, it seems to not like when advertisers and publishers innovate. It’s like saying “…wait a minute, that’s our job. Don’t go getting cocky about doing this yourself now that is a BIG mistake.” Maybe. Maybe not. Not all will win but that too is part of the ad tech cycle.

    The easy answer to all of this is transparency, which like porn, you’ll know it when you see it.

    The in-house trend will evolve into the marketer’s and publisher’s understanding of the true value of their own data. Once they have their data under their control, and can see across everything they own and that matters to them, then they can spend more effectively on any platform they choose. Advertisers and publishers live in the communications industry. Everyone in the middle lives in ad-tech. There lies the fundamental disconnect and likely one of many drivers of distrust.

    This is but a moment in time in the evolution of online advertising so if the “in house” trend drives another wave of innovation and creates a secondary movement to support their efforts, this trend could turn out to be the best thing to happen to the industry.

  5. I think that if you don’t do “in-house” programmatic buying as an advertiser, you’re a fool. The marketplace is too complicated and the opportunities are too great (I’m an optimist that believes advertisers can replace a majority of their online ad buys with programmatic spending) to leave it to anybody else’s hands.

    You may elect to have your agencies run a lot of the day-to-day activity, but one needs to have deep expertise in-house to direct the activity. In every experience I’ve had on the client side, the more time I’ve spent in the detail, the good questions I’ve asked–to the benefit of performance.