As we continue to improve our skill levels as marketers, it's essential to recognize the areas where we might get caught by surprise. Here are some tips for how to spot and serve for the mobile data curveball.
We've come a long way with audience-based planning and buying, beginning with more accurate audience definition, propensity modeling, targeting, and optimization over the course of any given marketing execution. Being on top of our game is all golden until you get a curveball. As marketers, and the partners and agencies supporting us, we've got to know a curveball when we see it and know how to solve for it.
Curveball #1: Knowing How, When, and Where to Accurately Identify Your Target Mobile Consumers.
This ball has practical and technical implications. At the beginning of any initiative, when we say we understand and then articulate who our customers and prospects are to anyone trying to help us with our marketing, we tend to think and express the definition in descriptive terms. Then, we strive to find data that consistently aligns with those descriptions. This description always includes obvious things like age, family size, living in a particular geography, and having distinct interests, when it comes to purchases or lifestyle.
But data that aligns with our descriptions, and data that aligns with our customers, and who they truly turn out to be, as a campaign plays out, can be different. This difference exists at the point where our expectations and true outcomes diverge. And such a divergence is likely to crop up when we move into a newer channel like mobile. The moment we spot this divergence of what we believe and what is really happening is the curveball.
What's happening in practice here is that the data available for marketers' use is collected through various channels and comes from a variety of sources that are compiled, self-reported, or based on surveys. These settings and sources are reflected, whether intentionally or not in the inherent resulting data values. When those data points are used in new channels, where you have not marketed before, such as mobile, they may or may not apply, and drive outcomes, in the same way we expect.
Imagine that you are a mattress manufacturer, and you have a new luxurious and feature rich mattress set, which you've found to appeal to a niche market. You've been able to define your target audience purely because you know your existing customers, with very detailed data points that describe things like, "I'm in market to buy a new mattress," and/or, "I'm a busy mom," and, "I enjoy home décor sites."
Further, you even design the perfect creative unit, with an attention-grabbing line or visual and a compelling offer. Let's even assume you've found a way to "know" the person receiving your message matches your provided descriptions. What if those descriptions don't apply to the mobile channel?
Though we've surely said it about every emerging channel, mobile once again delivers a unique and often initially unpredictable profile swing. This mobile-specific profile, and thus demand for a new way to accurately reach the mobile consumer, comes from two intersecting observations:
- The description of the mobile consumer population doesn't match the description of the overall consumer population. A simple illustration of this is that while your total customer population might be 20 to 60 years old, your mobile population is almost always going to be younger (at least for now).
- The general consumer profile or persona doesn't match the mobile consumer persona. For this example, look no further than the mattress maker story above.
With the very descriptive and well-defined profile in hand, applying it to reach busy moms, on home décor and related sites, on mobile, although your audience could very well fit that description, it's equally possible you could miss them altogether with this strategy.
Why? Well, suppose it's the middle of the day, and this target audience isn't in "mom" mode while on her smartphone, because she's also a business professional, and that's the hat and mindset she has on while using her phone. Suppose also her interest in décor sites is an evening relaxation thing she likes to do at home on her 21-inch monitor. In other words, while not captured by our initially described target, this ideal consumer is still available to be reached throughout the day, as she uses her phone while standing in line to pay for coffee, checks news, stock tickers, and sports scores, and admittedly takes in an occasional gossip story, or plays a game, for that needed five-minute mental break.
Your target audience descriptions, no matter how well culled, or even accurate, when not aligned with thecurrent channel, can result in two things: one, you may be in front of your target prospect, but you have the wrong message for their current mindset, or two, you never get in front of them because you're looking for them in the wrong places at the wrong time.
So, rather than regarding this discovery curve as unfortunate, as you might be inclined to do, when you realize your former descriptions are inadequate - you might realize the power of this level of discovery. The key is to work ahead with your team to develop meaningful audience models that incorporate an understanding of mobile behaviors, and allow for constant learning and adaptation of those models.
Curveball #2: Knowing the Person Receiving Your Message Matches the Description of the Inventory as Sold.
The opt-in world aside, it has always been, and will continue to be, a challenge to know the actual human being receiving your digital message is the person you're hoping to reach. Desktop digital can fairly effectively get to an understanding of the household, but still must solve who in the family or group of users is engaging, or must overcome a significant data gap when otherwise relied on cookies are turned off. Mobile on the other hand is almost completely 1:1 (people don't often share phones), and thus the curveball is that it's easy for the mobile audience to not act or look like you expect. And certainly, the audience won't act like users in a desktop environment. The marketer's pending benefit in this very personal and 1:1 channel, is the opportunity to make individual engagement decisions, assuming you have the ability to truly know who is on the other end of the smartphone screen. This can be achieved in a couple ways.
For one, we rely on real-time behavior as one way to effectively target in a new channel, without leveraging lots of previous data. We use technology that can learn in the moment. New technology systems allow the machine to learn who is doing what and what they are not doing. This data of the moment is incredibly powerful when factored into your models.
In addition, marketers would be wise to find a partner who has the precision and ability to capture and compile individual mobile insights at a device-by-device level, and also has the knowhow to apply those insights in an effective but privacy sensitive manner. A number of smart companies are building both cookie-based and cookie-less recognition systems to solve this problem while keeping both the buyers and sellers in mind.
And the best way to nip the curveball of surprise audience behavior is of course to do both of these things: keep a beat on mobile moments and partner well to zoom in on privacy compliant device-by-device analysis.
Curveball #3: Not Knowing If the Person Who Matches (and, in Fact, Proves to Be) Your Target Customer, Is Also Mobile Ad Responsive.
The background insights you need, in order to understand how to define, and in turn distinguish, your target audience, are the same insights you need to gauge responsiveness. You have to know who is in the mobile population, who is getting your message and not responding, and who is getting your messaging, and in turn engaging in meaningful and attributable ways. Get caught without this understanding, and you've just been whacked by curveball number three.
Direct marketers excelled at this idea of understanding message responsiveness. They knew that it wasn't enough to know if they were reaching their right target; they also needed to reach a responsive subset. Of course, they also had the advantage of easy access to name, address, and phone files. They also were able to easily track the subset they sent a direct message to through one of these direct channels, and once the consumer took action or not, they could thoroughly analyze the responders and non-responders - and then used this predictable and distinguishing information to define who they marketed to the next time.
Email marketers picked up where offline direct marketers left off, employing the same type of predictive response analysis for determining when, who, and how often to deliver an email communication. Of course, with email, exists the luxury of receiving response information in real-time versus in weeks.
Mobile marketers can and should be doing the same, especially in so much as this channel offers for the first time, the reality of individual 1:1 recognition, personalized messaging, and real-time response. Further, because it's more personal than desktop, and more disruptive than email, it can't and shouldn't stop with simple observations and predictions based on clicking ads. Mobile marketers should employ precise identifications of who is on the other end of the mobile device, what distinguishes them uniquely, what is relevant to them at the time, and what kinds of messages they'll appreciate receiving when. After all, we're boldly throwing our unsolicited message out there on their tiny personal screen and kindly ask for their time. The experience is better for all parties when this is done with insight and skill.
Of course in any given ballgame, there are more than three curveballs. But, you get the idea. As we improve and master our skill level, it's important to recognize a few areas where we might get caught by surprise or unprepared. But as data and technology systems evolve and allow for smarter preparation, we can work ahead of those surprises, and treat them as opportunities to learn and adapt.