Having a Fool for a Client...Why In-House Programmatic Is A Bad Idea

Yesterday I was speaking with a couple of colleagues and the conversation turned to the hot topic of Programmatic Buying and whether not companies should begin moving it in-house.

My colleagues both felt that it made sense to move in-house for a couple of reasons: First, there would transparency on the buying costs, which is still an issue for a number of trading desks and by some estimates put the "lost" client budget due to unseen fees or trading profits by as much at 30%. And, secondly, the company would "own" its data as opposed to being into a pool of data to be used by the trading desk and all their other clients, and in the era of Big Data nothing is more valuable than the data you own and can sell or not have be sold to others.

While we were talking I was reminded of Lincoln's quote about lawyer's representing themselves -- that the lawyer who represents himself has a fool for a client.

I was reminded of it because moving in-house seems to be similar mistake in many ways.

Why? Because it is inevitable in companies that control their own buying and data that they begin molding these to and for the company's agenda; in other words taking their own advice. Soon company's begin finding what they want to find as opposed to what the need to find. And they inevitably fall behind the market on modeling, pricing and best practice.

Let me explain by way of an example. A number of years ago I had a client that I and the rest of the industry recognized as one of the paragons of data collection, remarketing, prospecting, upselling and so on. In fact both Forbes and Harvard Business Review had covered their capabilities as "best-in-class". 

They had massive data base and six segmented audiences with at least three sub-segments for each parent segmentation and could pretty much predict when anyone would use their services and how they would do so.

When they had hired us they had reached a plateau and wanted ways to continue growth and had wanted media planning an buying to help with that. From the first meeting on it was apparent what the issue was and it wasn't solely having to do with media planning and buying. 

The company's very sophisticated segments and models were so sophisticated that that a data narcissism had taken over such that their model and segments were built solely on how customers interacted with their company, not on how customers interacted with the category nor the competitors nor the rest of the marketplace.

Now this might seem like a "no-brainer", however, I have seen this over and over again at companies that are considered, and are, best in class. They solely parse the data that is available to them and not to the whole of the marketplace largely because other data might not be available, or the costs are too high to buy it and they consider themselves the best anyway so what else can they learn?

Well a lot actually, and I think this is the most compelling reason not to move in-house. I think internal pressures and agendas are natural and very quickly the trading desk would be focusing on these. And, in the Big Data future being bound to one data set is going to impede growth.

It will be crucial to access data pools outside of a company's own existing database and go beyond narcissistic benchmarking and best practice regimens not only for media pricing but for enhanced modeling, prospecting, cross-selling, and upselling as well and one way to do that is to hire an outside Programmatic buyer.