Optimizing Toward The Target
Matt Arkin, senior vice president of advertising sales at Tacoda, has a bone to pick with the way BT campaigns get measured and optimized by too many planners. Raw click-throughs are not enough without a better understanding of how specific target groups are interacting with your messaging. Without expensive and lengthy post-campaign study, it is difficult to understand the branding effects of a BT campaign, and so he suggests ways of leveraging the native intelligence of a BT network to optimize towards specific segments.
Behavioral Insider: How are buyers evaluating BT campaigns? Rightly or wrongly?
Matt Arkin: It is such a wide-ranging set of measurements out there still to this day. The problem is people are getting away from measuring who they are reaching. So when the advertiser starts to measure campaigns and how things get optimized on the Web you are measuring just click-through. But guess who clicks through the most? People who enter sweepstakes, people that are looking for new careers. That is not necessarily the quality person that an automotive OEM advertiser wants to reach. From a behavioral standpoint, people who are affluent and in market for a new car, a Jaguar or Land Rover, are not really going to click at the same degree as people out of a job or the sweepstakes-oriented. So it gets down to quality vs. quantity of your target audience.
Behavioral Insider: What is at the root of that? Is the branding argument still not being made, or is it just too easy to optimize off of clicks?
Arkin: I think it’s expensive. I think it’s difficult to run an actual Dynamic Logic study. That tends to be one of the only ways to measure effectiveness of brand. We are starting to show that you can really move brand preferences.
Behavioral Insider: Through what kinds of metrics?
Arkin: We view campaigns through three different lenses. One is from putting our tag on an OEM’s Web site and looking at their organic traffic outside of a specific campaign. We are able to see what types of behavior segments -- affluents, parents, young and edgy -- and at what concentrations they come to the Web site.
Another lens is looking at campaign traffic and how it compares and contrasts with the organic traffic. The third lens is when people get to a certain point of a Web site, like building a car and getting a quote. So there we look at people who fall off, and what types of people aren’t getting to the quote stage, and how valuable are they?
So what we do is compare and contrast those three lenses and really start to show where there are opportunities, not only from a creative standpoint and messaging standpoint. We can start to show them where things are sticking and what types of behavior segments. Did the parenting- and family-oriented behaviors really drive through the funnel, or did they just get to a stage where they may be reacting to the ad, but didn’t drive towards the end of the quote stage.
Behavioral Insider: How is the data actionable, then?
Arkin: For example, the advertiser was an SUV maker and targeting that exact segment of family, parenting types of audiences. We were able to show that they reacted extremely well from an index perspective towards the creative. They clicked and interacted with the creative. However, we showed that they did not drive through the sales funnel to the quote. They [were lost] post-click and post-interaction. The behavior segments that were being driven through the funnel were more towards the gadget geek and electronics types of behaviors. A lot of it had to do with [the advertiser’s] message in the creative and how they pulled the client through the actual Web site itself. It was not a very family-oriented, compelling story on the Web site.
Behavioral Insider: How much time do you have to optimize off of data like this?
Arkin: We will typically request at least 30 days to optimize. If you are looking to optimize against clicks or conversions, we still do that. But I think, more importantly, we’re looking for 30 days to give you a full analysis in terms of what types of behavior segments are reacting, giving you some insights on where you have opportunity to target other specific types of behaviors. We can still show what behavior segments are producing the results for you.
We just feel that in order to optimize, you need to be open to the fact that it’s going to be a quality vs. quantity game. You have to optimize toward the best and most qualified audience. We have a two-by-two grid that shows how well specific audiences are reacting to your advertising. We report into how well your primary campaign metrics have moved the needle. The companies can change their messaging and we can show audience segments moving along that map based on changes to the creative or the Web site to that segment, which we think is the most important,
Behavioral Insider: What trends are you seeing in BT from agencies and media buying?
Arkin: We’re seeing much higher people coming to the table with us and more clients calling us directly. One of the problems with media buying right now is the classic musical chairs at the agency level in terms of who is doing the buying. There is very little consistency there. We include the agency with everything we are doing, but we’re doing an awful lot of client direct work. There is nothing more frustrating than getting a lot of RFPs for small buys. We just say no. We provide so many insights to a client that are extremely actionable.
A lot of times if you are dealing with an agency person, they don’t have time to do any post work. We’re saying no to business more than we say yes because it’s either too small a commitment of time, resources, and money, or it’s just at a level where we don’t think they will concentrate on the back-end insights. In order to do this the right way you need a fairly large commitment -- not just money, but in resources on the back end of the findings.