Where does programmatic data come from?

Kyle Minerley

As we’ve mentioned before, programmatic advertising is all about targeting. We serve up ads for the right audience at the right time in the right (digital) place in order to reach customers more effectively and ultimately get a bigger return on ad spend.

In order to target effectively, we need data. Here’s where it comes from:

Inferred Data

When we say “inferred” in this context, we’re talking about drawing conclusions about users based on their browsing behavior.

One of the main programmatic data sources is cookie-based analysis of web traffic. If we see that a particular user frequently visits golf websites, or the golf sections of general sports websites, we can reasonably understand that this user may have an interest in golf. That’s a simplistic example, but the more data we have to analyze, the more sophisticated inferences we can draw about people using lookalike targeting.

As the internet goes mobile, another key data source is mobile app data or mobile device data. We look at where people are physically going, which apps they have on their mobile devices, the genres of music they listen to, and so on. A mobile app developer, for example, may target people who tend to make in-app purchases because they fit the profile of a high-value customer.

None of this inferred data is tied to any personally identifiable information. We just use it for lookalike targeting.

Declared Data

Among the biggest players in the data space are major credit card companies like Visa and Mastercard. These companies have a tremendous amount of purchase data because they can see and track anything that is purchased with one of their cards, anywhere in the world. They also have access to information such as available credit and overall spending behavior using credit cards.

On the flip side, the individual retailers, whether brick-and-mortar or digital, have a vast amount of data on purchases made at their stores. They know where their customers are located, how much of each product they buy, whether they return or keep products, and so on. This type of data includes more granular insights on spending patterns within particular categories.

Data from these sources is more adjacent to personally identifiable information, but it’s still scrubbed of anything identifying an individual before we use it for targeting.

Data Syndicates

Think of the syndicates as a Venn diagram, looking for relationships and commonalities within particular sets of data.

To return to the previous examples, the credit card companies have a great deal of information on spending with their cards, as well as available credit and other closely related information, but they can’t track purchases made with cash or checks. Meanwhile, the individual retailers have a great deal of information on purchases made at their stores regardless of payment method, but they have no insight on purchases made at other retailers. By cross-referencing and overlapping these two data sets, a data syndicate can start to draw deeper inferences about overall spending patterns and target customers more precisely.

When you get into the weeds of data syndicates, there’s a surprising amount of data sharing even among companies that are nominally competitors (think of Coke and Pepsi, for instance). Ultimately, having better targeting serves everyone’s purposes.

Inventory Control

To shift gears from targeting for a moment, we also use a lot of data for brand safety purposes. There are several data companies that compile information on which websites pose a risk to brands that advertise there — think of the difference between advertising on a mainstream media site versus a site with extreme political views, or a mainstream business versus an adult website.

Of course, there are brands that are perfectly ok with advertising on certain high-risk sites. The goal is to have the tools to create a brand safety strategy that makes sense for each advertiser. We use data to enable brands to remain socially conscious and target low- or medium-risk places if that’s what’s required.

In the end, knowledge is power… literally

There’s a reason data is emerging as the world’s most valuable commodity. When you have the right information to target your ads precisely while protecting your brand, you’ll see your return on investment skyrocket. If you’re ready to put the power of data to work for your business, contact us today to speak with a THRONE expert.

* Denotes required field