In the world of digital advertising, there are few terms that sound so simple but we struggle to explain. Programmatic is a good example of a concept that is widely used in our industry but rarely do people understand its true meaning.
Read other posts from Multi-Solution Architect Blog Series.
Let me set the right expectations for this post: I will not be covering the history of programmatic, how it evolved and where it is going. There is so much detail on this topic that we will need a book to cover it entirely. Instead, I will be concentrating on the architecture of the programmatic space and its most valuable currency, data.
First things first, let’s define programmatic. The way I envision programmatic is nothing more than an automated way to buy and sell inventory. Wait a minute, we thought this was about advertising. It is indeed but what is the value of an Ad if you cannot display it anywhere for people to see it? Think of this as the real estate model where you have a buyer, a seller and a house. Mapping this to programmatic: the buyer is the advertiser; the seller is the publisher and the house is the inventory except in digital marketing you have millions of houses that need to be sold and to be occupied with a blink of an eye. The ancient way of the advertiser calling the publisher and negotiating the offers on an inventory doesn’t scale up with this volume of inventories and hence why an automated way to handle the negotiation, selling and buying is what makes programmatic a no brainer.
In order for this to work, both parties need a mechanism to set their own pricing, terms and rules. In real estate, this is usually represented by realtors where each party will have their own representing them and their properties. In programmatic, Demand Side Platforms (DSP) represent the Advertiser (Buyer) while Supply Side Platforms (SSP) represent the Publisher (Seller).
So what does representing the buyer/seller mean in programmatic? For advertisers, the DSP enables them to set the following rules:
On the other hand, the SSP enables the Publishers to set the following rules of engagement:
A common space for the DSPs and SSPs to play is the Ad Exchange Market. The Ad Exchange is where the Publishers release their inventory along with the rules they set up in the SSP and where the Advertisers fish for audiences defined by the rules set up in the DSP.
The complete story of how this space operates including private exchanges, real-time bidding, ad networks, ad servers, trade desks and more goes beyond this post and we will save it for another time. Instead, let’s talk about the role of data in programmatic advertising.
What data are we exactly talking about? It is the data that informs the advertiser that there is a good deal on the table. Let’s look at this from a different angle:
Lovely story, right? Is it clearer now that the data that defines the visitors is what enables the programmatic buy and sell? Yes, it is the Audience Segments/Attributes.
Audience Segments can be defined based on 1st, 2nd and 3rd party data sources. They can be a combination of behavioral data, demographics, personas, partner data and more. These audience segments exist in all shapes and flavors and are fed into the DSP, SSP and the Ad Exchange. In most cases but not necessarily all, a data management platform (DMP) is the bank that hosts these audiences and distributes them to the players in the programmatic ecosystem.
The data that comes out of the programmatic advertising is equally as important as that data that feeds into it. We are often occupied with finding the right data to feed into the programmatic space. Programmatic is able to process billions of transactions a second so imagine the volume of impressions, clicks and conversions that are generated. This data definitely makes beautiful reports but how about the learnings and insights that it produces?
Here are three ways you can utilize this data:
Programmatic Advertising is continuously evolving and introducing new platforms such as TV (Linear and Addressable) and Out-of-Home which means that we will never have a shortage of data or insights to work with.