When it comes to reaching consumers with mobile advertising, is there really a difference between an Apple iOS user and Android users? A poll commissioned a few weeks ago by mobile real-time bidding operator AdTheorent suggests that there is. Conducted by Quinnipiac University, the survey found that while Android-based smartphones were the most widely used mobile device (39 percent) followed by iPhones (30 percent), iPhone users were relatively young – 80 percent were under 35 years of age – and according to AdTheorent, they’re more receptive to mobile browser-based ads.
We spoke with AdTheorent CEO Anthony Iacovone about the differences between Android and iOS conversion rates and how the company views the promise and problems associated with mobile RTB.
How did AdTheorent form? And are you solely focused on mobile RTB?
ANTHONY IACOVONE: Yes, we’re a mobile RTB pure play from the ground up. AdTheorent was born from my last long gig, which was at Augme Technologies. We bought Hipcricket, since we’d seen that media properties were getting really hot.
However, Augme didn’t have a really big appetite to get into the media space, so this was where the break point was for me. Another thing we saw – this is about a year-and-a-half ago — was that RTB was just taking shape. But there were still uneven gaps that we felt we could close.
What are the problems in mobile RTB that you’re trying to address?
Primarily, the problem is that everybody’s measuring off the click. In our mind, the click doesn’t mean everything. It just means you’ve gotten somebody close to where you want them to be – but questions remain: Did they actually convert? Did they actually buy something?
We spent a lot on building post-click conversion tracking for our advertisers within their apps and within their sites on the mobile web. In the middle of this, is what we call “RTOM,” which is a real-time learning machine. It’s all predictive modeling. It came out of the recognition that old school data mining models don’t work extremely well with big data.
Why is that?
Digital advertising involves an enormous amount of data, 30- to 50 billion ad requests, responses, appending data, and lots of variables. When you start modeling on that data and something changes, you have to redo your models. It can take several months to redo a model and get it right again.
Our RTOM solves that by actually having a self-learning system. Within real-time, the model can morph itself based upon any changing data — adding data, removing data. It’s very flexible. It keeps it soft and we see, versus a random set, 200- to 300 percent lift in conversion using that model.
As a business, RTB is still pretty young and mobile RTB is still in its infancy. So why focus on closing the gaps in a business that still has a ways to go in terms of scale?
The straightforward answer is that the mobile RTB space does currently exist as a primary buying channel for media agencies and the brand directly, if the brand chooses to go that route. The impression base is enormous. We have reached the threshold of 30-to 40 billion impressions on a monthly basis. We can see today, there are a lot of exchanges that have gone and taken their inventory and made it RTB-enabled.
So what is the chief hurdle in the advancement of mobile RTB?
The hurdles are around education. Heads start to turn a little bit when you start to say, “First, RTB.” People hear it, but they really don’t understand it. You have to take the media planning agency, and you have to walk them through it. I’m talking even at an individual planning level. You have a lot of great, young media planners who are willing to learn, and this is the first time they’re hearing it. We spend a lot of time letting them understand what RTB is, why you buy one impression versus another. It’s not only from a rate perspective, but also from a targeting prospective, and awareness campaigns, from a conversion prospective in a direct response campaign.
When you’re buying – and I don’t want to mention names – first gen network, the big names, the public companies, you’ve got to spend the time with them to let them understand that you’re not just another network; that you’re something completely different. That’s a big hurdle.
Technologically, I can’t see any hurdles that get in the way. Today, the processor speeds of the servers that we use, cloud-based computing, can make decisions in milliseconds. The way we can enrich data in milliseconds—it’s all there. If you asked me six years ago, it’d be a different story, but today, it’s all there.
What kind of advertisers are interested in mobile RTB? Any categories of marketer in particular, or is it pretty general?
I was the first investor when we started AdTheorent and I’m also its operator. The first thing I wanted to see from the business as an investor was big brands doing some awareness campaigns.
You’ve got the big e-commerce companies spending in this. The verticals that have been stronger are all of the same ones that you’d see strong in traditional digital with us, so we’ve got huge awareness campaigns with some of the biggest banks in the U.S. The same is true when it comes to verticals with luxury brands. We’ve just started to penetrate automotive.
When it comes to mobile, Apple’s iOS and Google’s Android systems are dominant. Which system is better for mobile RTB?
Our numbers show that iOS has a better conversion rate when we see a DR campaign. We see less value out of Android, but we can drive the rates down. The rates are typically lower there, so it’s equitable in some sense. Apple’s iOS really performs better. If I were looking at it from psychological perspective, I’d say iOS user appears to be somebody engaged as a primary browser.
It appears the Apple user seems to be more engaged with that device as a primary browsing device. And if you’re engaged as a primary browsing device, you’re going to engage with the advertising as you would potentially on the web.
