The Four Stages Of Mobile Marketing
Mobile marketing success is dependent on the degree in which you measure, analyze, and act upon the wealth of data at your fingertips. Which of the stages outlined in this article best reflect your relationship with mobile data, analytics, and measurement?
It doesn’t matter which app you’re promoting; one thing is certain: It’s creating tons of data pieces all the time. The question is what you’re going to do with all this data and how you’re going to build the puzzle.
According to some pretty shocking research considering we’re in 2016, no less than 50% of marketers still aren’t measuring user engagement and mobile ROI.
Robust mobile campaign measurement is crucial as it directly correlates with sophistication, which should correlate to success (assuming you’re doing things right and following the data trail). Despite the challenges of cookie-less mobile measurement, there are ways to overcome this and accurately leverage data to take your app promotion campaigns to the next stage.
Which of the following stages best reflect your relationship with mobile data, analytics, and measurement?
Stage #1: Measuring Installs
An install is an important milestone in the app marketer’s funnel. But it’s only a stop on the way and definitely not the final destination. Since the number of installs is an important factor in app store optimization (ASO), its main value lies in driving growth of organic installs via higher app store rankings. Many companies offer ASO solutions with each having its own experience and different approach for each store.
Tracking can be done by adding the SDK (a piece of code that enables the gathering of app data) of every ad network you run with, in addition to the SDK of an in-app analytics provider. However, if you’re running with multiple networks, too many SDKs will slow down your app, which is something any app developer should not tolerate.
In such a case you should work with a mobile measurement company—here’s a list of those vetted by Facebook—that has a single universal SDK that is connected to hundreds, if not thousands, of partners. This enables a bird’s eye view of the ecosystem to determine accurate attribution, in addition to measurement of post-install events to understand their real value.
However, the problem with measuring installs in an app environment dominated by the freemium model is that it has little effect on generating actual commercial value. That’s because most users who install an app do not actually use it over time. Either it remains buried in their phones somewhere or they simply uninstall it.
The following table illustrates my point:
As you can see, network two drove almost twice as many installs as networks one and five, but when it came to actual value, it only produced 95 paying users. That’s a 1.15% install-to-purchase conversion rate. (Network one’s rate was almost six times that). Network two also delivered a very low $0.24 average revenue per user. If you only refer to the install column, you’re wasting a lot of money.
So if you’re only measuring installs, you should move on to stage two.
Stage #2: Measuring Basic Post-Install Events
This stage usually arrives once the understanding sinks in: You need loyal users that drive real value. You want them to revisit your app, replay, and hopefully rebuy. At this point you don’t really know what got them to come back, and you don’t have audience lists.
You can measure app launches to understand which source delivered loyal users. You can also understand which events are commonly used—and if you’re a gaming developer—is it tutorial completion, levels passed? If it’s a shopping app, which products were viewed and which were added to the cart? There are several ways to measure in-app events by adding them at the SDK level or through a server-to-server integration.
By tracking the events back to the marketing channel that acquired them, you can pinpoint which channel, and also which media source, campaign, publisher, or even creative produced the most relevant audience, and then spend more (or less) with that specific source.
Stage #2 requires some effort as it’s an exploratory stage where you’re still feeling your way in the dark. You have to identify the in-app events you need to measure that will help you meet your key business goals. In addition, you need to pinpoint which sources are able to deliver users who meet your goals and which creatives on which networks deliver the best engagement.
After mapping events that are relevant to your business and measuring them, you can use retention reports to find out which source delivered loyal users by monitoring app opens on days or weeks after an install. The retention table below clearly shows that network one should be dropped, while network eight should be mimicked.
To gain a deeper knowledge on your audience and the marketing channels you work with, use cohort analysis to understand the quality of your average user over time. This is done by grouping people with common characteristics and measuring specific key performance indicators (KPIs) over different timeframes. It’s therefore a really good indication of change over time.
The following graph shows a cohort of users from the U.S. who installed an app during October:
What Do We See?
- Media source A starts off with a bang, producing high quality users who are highly active during the first 14 days after which the growth rate begins to plateau.
- Media source B is also delivering quality users showing a modest yet steady climb throughout the period.
- Media source C is slowly growing until day seven, but then flattens.
- Media source D performed poorly. Not only did it acquire low quality users that barely generated any revenue, it also did not show any improvement over time.
What Can Be Done?
- Having noticed the drop on day 14, the marketer began retargeting his users at that time to encourage them to continue engaging with the app.
- The budget of media source B was increased hoping it would generate more users who would only increase their engagement over time.
- Media source C was broken down to the campaign and ad level and those that underperformed were removed.
- Media source D was removed.
If you’ve completed this stage, you’ve made a significant step forward in your app marketing!
Stage #3: Measuring Rich In-App Events And Retargeting
This is the point where you understand that there are five to seven in-app events that are crucial to your business. You focus on them by breaking each down to a parameter level. So instead of measuring users who added a product to the cart, you’re measuring users who added at least two women fashion items priced above $50 to the cart. Or if you have a gaming app, you measure users who completed a game tutorial and viewed the third tutorial, registered with Facebook, and went on to clear level five.
Armed with such deep knowledge of your audience, you build highly segmented audience lists and run personalized retargeting campaigns in order to better communicate with those users and maximize their ROI. You also look to enhance your user acquisition strategy by targeting “look-alikes,” understanding that these potential users will yield roughly the same ROI as your current profitable users.
This stage is really important as it helps you generate continuity—driving users to continue where they left off. For example, encouraging a user to return and buy the piece of clothing he viewed but never bought or reactivating a high-value dormant user by offering a discount on a product he’s likely to be interested in. This can be done by measuring—through the customer user ID—the revenue of a user’s purchases to determine value, the stuff he’s looked at to pinpoint products of interest, and the date to determine how long that user has been inactive.
Stage #4: Measuring Cross-Channel (As Opposed To Multichannel)
You are a data beast, measuring across channels and realizing, to the best of your ability based on what can be measured, which role each channel plays in the consumer journey—across devices and even offline, most of the time. That’s because a complete solution that connects all the dots is still not possible. Even Facebook or Google who have hundreds of millions of users who identity themselves across devices are limited only to, well, Facebook and Google properties, respectively.
The good news is that we’re heading there by leaps and bounds. Cross-device measurement is already possible by using deterministic identifiers like email and probabilistic solutions that use big data to predict the likelihood that a user with tablet A is the same user as smartphone B. Since it’s a statistical model, however, it doesn’t offer bulletproof accuracy. It’s usually very accurate in the short term and then drops over time. (Click here for more on cross-device measurement.)
The following illustration shows how a properly connected campaign can completely change a marketer’s conclusion:
Cross-device/omnichannel measurement is perhaps the number one challenge for marketers today. This is driving innovation in the mobile ecosystem, which in turn closes more and more loopholes.
The bottom line is that mobile marketing success is dependent on the degree in which you measure, analyze, and act upon the wealth of data at your fingertips. If you’re at stage one, you’re probably wasting a lot of money. As an aspiring marketer, you should always look forward, one step at a time. So start measuring post-install events, and then go deeper and deeper to squeeze the data lemon to its fullest. When you do that, you’ll get a great-tasting, eye-opening fresh lemonade that will keep you asking for more.