Associate Media Director
Kyle Lebet is an Associate Media Director at 22squared. He has worked in digital media since the good old days of 2008, after a short stint in spot broadcast. Even though he rides a bike to work, he has worked almost exclusively within automotive in that time, first at Mediacom with VW and then at 22squared with Southeast Toyota.
For us that aren’t familiar with 22squared, tell us a little more about your agency, the clients you represent, and the core offerings that you provide to your clients.
22squared is a fully independent, fully integrated agency representing national and regional clients across all marketing disciplines. We strive to take full advantage of that integration, bringing strong storytelling and engaged thinking to the table while also maintaining a cutting edge digital technology suite to support media and analytics and make sure those stories are heard.
When it comes to audience buying, what are some strategies you use when trying to most effectively reach this audience?
It all starts and ends with data management. Since this is led by brands on many cases, it is a matter of maintaining very close relationships with those in non-traditional marketing roles, very heavy CRM- and data-focused, in addition to the usual marketing team. There is so much valuable data within the brands we work with since they do a great job in knowing who their customer is. It then is incumbent on us to translate that data to actionable insights and reachable segments.
Tactically, do you have allocations by platform to reach this audience, or do you view them holistically in a channel agnostic way? How do you prioritize channels during audience buying?
We must prioritize by channel in many cases for budgeting purposes, but as soon as we start seeing audiences pop up one place or another those allocations shift quickly. So we are striving for that agnosticism, but we have to start somewhere.
What are some of your biggest challenges when it comes to audience buying?
There is still some separation by discipline. Video and Social can occupy their own silos if you are not careful. Also, since more accurate targeting by log-in based sites is the future of this, figuring our how to get the data from those sites more closely aligned with advertisers’ site data.
Do you make use of a DMP to integrate any 1st party data from a client, and how do you go about doing this? Can you give specific examples of how you leverage 1st, and 3rd party data sets to more effectively target the optimal audience for your client?
This varies so much by client. Since DMPs have applications far beyond marketing, brands will often manage this relationship directly. That means we are working with brand data marketers to suss out what exactly we have access to, and can target and learn with. But CRM data is the cost of entry here, and so however we go about the DMP relationship, there is 1st party data as a baseline. 3rd party is a little more complex, as its accuracy is questionable in many cases, but it can still be useful as a double-check on scale and performance.
They are not really decreasing. They will be managed differently, as we will be deciding which impressions to serve and when, but these will mostly still be bought and negotiated upfront. It’s not about just getting cheaper impressions, it’s about making the impressions you already buy more valuable by controlling the targeting and frequency.
What do you envision might happen to ad networks, both horizontal and vertical, over the next few years?
They will still have value as long as they perform. Since most ad networks are really just managed-service DMPs, if they can prove that the math works then they will have a place. As far as more specific vertical networks like Sojern or even more curated like Jumpstart, these were never really like an ad network in the sense that they were always more premium and niche.
Where do you see the role of premium publishers, as it pertains to direct relationships to advertisers, in a world where technology seems to be commoditizing everything?
There will never be a way to programatically buy a deep sponsor relationship. Media brands with cultural cache will still be able to commoditize and sell it in analog.
How would you personally define programmatic?
Serving ads on an impression by impression basis. Not buying ads, which would be RTB, but serving.
GroupM announced in Q3 of 2014 that they were going to cut all open exchange buying by 2015. What are your thoughts on this, and what effect do you think it will have on the industry? Do you think there will ever be a point where most of the direct buying is done through private exchanges, or Deal IDs?
It depends on whether we see further consolidation of audiences on a few big platforms like Google, Facebook, etc. Otherwise you will never be able to find your audience without resorting to open exchanges. The decades-long big-bang of audience fragmentation has begun to reverse under platforms like these, but until they are a little more open to integration with other platforms then, open exchanges will be necessary. I think it’s safe to assume that these open exchanges are also going to try very hard to ameliorate some of the issues GroupM expressed in the meantime.
What is your take on audience buying as it applies to viewability?
Viewability is sort of a myth. No impression is inherently viewable or not until it no longer exists. The key question then is, what impressions are you leaving on the table by hewing to unreasonable viewability standards? If that below the fold 728×90 performs at a 5x level, but is only viewable 30% of the time, it is still a good impression and should be served. A viewability prediction metric will be a huge number to include alongside all the thousands of data points that go into decision engines, but it will still be only one data point to consider.
Over the next few years, a lot of brand dollars will start shifting to mobile; how do you think advertisers will tackle the cross-device issue? What has been your experience with the platforms with registration data, and those with more probabilistic methods?
Platforms like Facebook, Pandora, Google, Instagram, et all have such a huge head start here that it is difficult to imagine them losing out. They know who users are across devices sort of automatically. They also have massive reach, which is leading the consolidation we talked about earlier. Then you have considerations like the AOL-Verizon merger that could really tie in hardware, so that will be transformational as well if you see more actions like that.
Lastly, what do you see as the future of audience buying?
Well broadcast is the obvious holdout here. There is robust debate internally at 22 and elsewhere about whether this is because the technology isn’t there or because the broadcast giants don’t feel like splitting up their audiences. It is much easier to make lots of money when you sell audiences in segments of millions instead of segments of 1. The reality is likely somewhere in between, in that the technology may not support full-on addressable TV but that’s only because the investment isn’t yet worthwhile. One day there will be enough people angry at their cable companies to swing that I’m sure. Just don’t know when.
22squared is an independent, fully-integrated advertising agency that has built human, inventive and inspirational brand experiences for nearly 100 years. As an award-winning agency, we help brands become Welcome Intruders℠ by creating ideas with scale that connect with consumers emotionally and drive powerful conversations. We support iconic brands, such as American Standard, Baskin-Robbins, Dunkin’ Donuts, GNC, Mizuno, Publix Super Markets Inc. and Southeast Toyota Distributors, among others with teams based in Atlanta and Tampa, Fla. Visit us at 22squared.com to learn more.