Mobile analytics: Measuring message impact for optimal ROI
August 28, 2013
Brendan O'Kane is CEO of OtherLevels
It seems almost incredible that, back in the mid-‘90s, marketing on mobile was limited to the 160 characters that could be used to compose text messages – no photos, audio or video.
That was followed by marketing through mobile email and mobile Web sites after the earliest massmarket “smart” devices were released in the mid- to late-‘90s. MMS surfaced in 2002, giving marketers a way to add some multimedia flair to their messages.
Mobile marketing made a giant leap in 2007, when the first iPhone showed both consumers and marketers the breadth of communication possibilities.
A year later, Apple pushed the envelope even farther by allowing users to download third-party applications through its new App Store.
Today, according to recent research, U.S. smartphone and tablet users spend a daily average of 158 minutes on their devices.
Furthermore, 80 percent of that time is spent posting to social media, browsing news and playing games on apps, which are their own universe of potential customer engagement and retention.
Considering this fast-moving history, it is crystal clear that, while it has not arrived yet, the day mobile becomes an essential part of every marketing campaign will be arriving soon.
Getting started is hardest part
This really hit home when I read the results of a May 2013 study released by the Mobile Marketing Association and IHS Global: mobile marketing spend will reach $10.46 billion this year and $20 billion by 2015 – in the United States alone.
We know that mobile marketing is on its way to inevitability. But there are still many marketers who are finding it difficult to pinpoint their first steps.
Some brands have not made the leap to mobile marketing because of their fear of being perceived as spam. Others are reluctant because they are not sure how to gauge the return on their investment.
For most of the mobile marketing history I referenced earlier, brands have reached consumers either through SMS and MMS on feature phones and through mobile email after the first smartphones came into play.
Two years after introducing the iPhone and one year after the App Store’s opening, Apple launched its Push Notification Service.
Now, as the feature phone’s previous dominance is being eclipsed by the smartphone and the average owner’s device has 41 installed apps, there is a whole new generation of in-app messaging that marketers are starting to leverage to better connect with customers.
For example, while push notifications originate on a server and are sent to customers’ devices, local pushes are triggered by a previous app action by the user and delivered to the user’s own device. These can be reminders of scheduled appointments or outings.
There are also app-originated push notifications that appear on a user’s device after being triggered by an action on someone else’s device. A good example of this would be an alert that tells a user it is his or her turn to make a move in a social gaming app.
Whether it is “to-device” or “on-device” formats, marketers can use mobile messaging to do a lot more than just drive sales.
For instance, relevant and timely mobile messages can also help brands engage and retain customers through fun gamified promotions, manage customer relationships and foster brand advocacy through socially shareable messaging.
Segmentation and A/B split testing
Before marketers start designing and testing messages, it is important to know who the target audience is.
It is also important to remember that audience segments are not static – they are dynamic and alive.
For instance, customers can quickly move in and out of segments – e.g., English speakers, Apple users, social sharers – and belong to several segments at once.
Segmentation can also be based on whether or not a customer has opened and acted upon – or not – previous messages and on how frequently a user opens an app.
Once the target segments are in place, the best thing a mobile marketer can do is run A/B split tests, creating two or more variants of one message to see which drives more of the desired conversion, be it a registration, a social share or a sale.
The beauty of A/B testing lies in its ability to remove guesswork from messages’ creative elements, actually enhancing creativity by allowing marketers to question their own assumptions and easily test hunches. Quick results make it possible to adjust messaging on the fly.
For instance, while testing push notifications for users of its mobile app, Australian parenting site Kidspot found that its audience of moms prefers shorter, clinically worded messages addressing stages of pregnancy. By adjusting messages accordingly, the site eventually saw a 200 to 300 percent lift in app engagement.
My next article looks at the outcomes of a split test and discusses how to interpret the numbers that come in to determine the winning mobile message, as well as how to use that information to retarget unresponsive customers for maximum engagement and ROI.
Brendan O’Kane is CEO of OtherLevels, San Francisco. Reach him at .