This month, Admap explores frequency and asks, how much is too much? GroupM’s Max Jaffe reminds agencies and brands that when it comes to programmatic advertising, the details really matter, especially from a measurement and frequency enforcement perspective. Rather than act on assumptions, agencies and brands are encouraged to ask questions and push for facts around what can and cannot be done. This will ensure everyone remains focussed on positive results relating to audience exposure control and frequency.
In this Q&A, Max Jaffe, Managing Partner, Programmatic Practice Lead at GroupM, offers his advice on managing frequency for programmatic advertising.
Do the old rules of frequency still apply on digital and is there an optimum number of times someone should see an ad before a marketer is wasting money showing it to them?
I think the concept of optimal frequency is still very relevant. Fortunately, we're able to measure more and take into account things like off-line sales and utilization of CRM data and more first-party data. This ensures we can have a holistic view on what a true ideal frequency is, whether it's driving an online type of conversion or potentially an offline one. We've seen a big uptick in a lot of brands, as well as our own efforts, focusing on how to activate and understand client's first-party data better in order to apply that knowledge in multiple ways. But a lot of it really is around activation and measurement, as well as discovery to better understand audiences.
Does frequency management for programmatic work properly?
It's a challenging question because the definition of programmatic-- or what we can do programmatically, is constantly expanding. There has been a lot of progress in the platform's capabilities to do this effectively. As more channels such as connected TV, OTT, digital out-of-home or digital audio become available within programmatic plays, there are new challenges that come with it.
What are the limitations for frequency management for programmatic?
Controlled management of frequency is definitely tied to the trend around brands wanting to have more platform consolidation, for multiple reasons. The main goal is to have control over their audience and control over exposure to what they're doing. We have some channels we can control such as the traditional desktop display video channels as well as a growing amount of mobile channels that fall within that realm.
The gap is starting to close on certain elements of connected TV and audience targeting capabilities which are starting to ramp-up.
With those emerging channels, it's really a combination of utilizing the tools and capabilities we have to get closer to the audience and then pairing that with campaign objectives. It’s important to set up best practices so we can control how much we're serving certain properties, to try to limit over-exposure when we can't directly frequency cap on a user level.
Do you expect this to transition into frequency management via addressable TV?
Yes, I think so. This is an area that we're diving into deeper. What we're seeing is a ramp-up of activation of addressable TV through programmatic platforms. What's being worked on, especially with larger companies, like AT&T and Verizon, is the tracking element and the ramp-up of more proprietary device graphs, as well as activation across owned and operated properties programmatically. As those device graph identity solutions become more widely available we'll start to see gaps close with targeting on the addressable TV side.
What's the timeline on that?
I think it's going to be a while before addressable TV solutions are widely available. This year and next we will see an increase in the volume of inventory and a lot more consistency around identity in that space and how we view it. Then new tools will start to flow in around strong enforcement of frequency management more holistically. We’ll make some progress in 2019, but I think in terms of true audience-level or user-level frequency, it'll be a little while longer.
How does frequency work between channels? For example, how realistic is it to manage frequency across multiple channels? Or do you have to take a channel-by-channel approach due to inconsistent measurement frameworks?
I would say our ability to manage the crossover in channels is definitely starting to close. When you're looking at a desktop-web environment we have a lot of strong capabilities. There has been a lot more control in the mobile app space, as well as in mobile web. Certain elements, for example within the Safari browser have changed in terms of identity control and user targeting. But holistically, we've been having a lot more success in our ability to control cross-channel frequency within our platforms. We have also seen the growth of capabilities like programmatic guarantees, even in a desktop display campaign, because it delivers higher access within a publisher's waterfall, guaranteed inventory and offers a new area where greater frequency control can be applied.
It has to be a balance between a publisher being able to guarantee those dollars, without a buyer restricting what impressions they buy. So overall, we are seeing a deepening relationship between publishers and buyers which is a really positive thing overall and enables us to discuss best practice around activation and audience control.
What mistakes are advertisers making when it comes to frequency capping and management in programmatic?
From a programmatic perspective, one of the biggest challenges that I've had historically, is programmatic being represented as a single line item within measurement solutions or analytic reports. When frequency analysis is taking place and you have all of these different tactics, formats or channels bundled into this single programmatic bucket it becomes challenging to truly gain actionable insights. A lot of brands and agencies still think of programmatic as a channel as opposed to a way to activate media.
If you think of a retargeting-type tactic, for example, where there's very high intent with users, a higher frequency makes a lot of sense. But retargeting could be paired with something that's more awareness driven, where you want greater reach. When those tactics are bundled together in the analysis, and you're being told what the optimum frequency is, it really doesn't give a truly actionable picture on a tactical o strategic level.
What we're focusing on is digging into deeper layers on the measurement side within what we're doing in programmatic or in digital, to get more precise detail about optimal frequencies.
What are the potential consequences of poor frequency management in programmatic among end users?
I think it can really impact performance at the end of the day. Not messaging the ideal amount of times towards those highly-engaged, high-potential converters, and possibly over-engaging within an awareness tactic because of that blend of different tactics all within one bucket.
What role could or should media owners play in ensuring brands better manage ad frequency?
I think there is this growing trend of exchanges or publishers having a much stronger sense of their user base. They’re also taking a more considered approach to how they're engaging with their audience. In areas like connected TV, or potentially the addressable space, where measurement is still a little bit weak on the targeting side within the DSP platforms, there’s a big opportunity to utilize those publisher relationships. To deliver better control and understanding across how often they're showing certain ads, to which users, or how often they're sending out messages within certain brands or brand categories, to help complement our efforts on the buyer side.
Our goal has always been to work closely and more strategically with publishers on behalf of our clients.
Will new technology such as ID management through Blockchain solve frequency management issues for good?
We have been diving into the Blockchain technology but we haven't really tied this to identity, it has been more transactional.
Ultimately, for GroupM, this is an education piece for brands and agencies, we encourage them to ask questions so they don’t assume nuances - especially within programmatic. The details really matter especially from a measurement and frequency enforcement perspective. I think it’s important not to act on assumptions but push for facts around what can and cannot be done to ensure everyone remains focussed on positive results relating to audience exposure control and frequency.