Mobile Engagement Crash Course #12: Advanced Optimization

By: Kahuna | June 26, 2015 | Mobile Marketing
Mobile Engagement Crash Course

In our last Mobile Engagement Crash Course, we discussed how mobile marketers can utilize personalization, timing, and location to optimize their campaigns. These are vital ways for marketing professionals to better deliver messages that resonate with users in the Mobile Era.

But these aren’t enough if you’re truly aiming to delight your users. In this post, we’ll dive into a few more advanced tactics to consider when executing an effective mobile messaging strategy. Deep linking and machine learning are two powerful levers mobile marketers can pull to further optimize campaigns.

Deep Linking

First and foremost, all of your marketing messages should protect, if not enhance, your user experience. With the exception of pure brand awareness messaging, you’ll also have specific goals you want the mobile campaign to trigger. Incorporating deep linking can achieve both of these objectives.

Deep linking enables you to direct the user to a specific place in the app, instead of just having the link open to the app’s home page. For example, the SeatGeek app could send a push notification about the Golden State Warriors with a deep link that goes directly to the portion of the app that handles Warriors tickets.

Dynamic deep linking enables even more personalization because it customizes that unique place in the app based on the user’s past behavior or preference. For example, if you’re trying to get a user to create a playlist in a music app, a dynamic deep link can send that user directly to the genre of music, or artists they’ve previously showed interest in.

Both of these examples do a good job of providing a better user experience because they remove friction from using the app. Clicking on a message without deep linking would simply open the app, forcing the user to navigate to where they want to be before the conversion event. Those extra steps are a pain for your user, and it’s a major drag on conversion rates.

We used the music playlist campaign as an example, but this type of deep linking can be applied to a variety of actions, including completing a registration, sharing a video, or performing any end goal your company deems virtuous. With deep linking and dynamic deep linking, marketers can further optimize their campaigns based on stronger engagement and goal achievement rates. Moreover, users are becoming accustomed to—and almost expecting—deep linking behavior, as iOS and Android are incorporating more of this personalization directly within their operating systems. As you dive deeper into your mobile messaging strategy, deep linking will become a critical tool for you to leverage in order to provide a more personalized and engaging experience for your customers.

Machine Learning

“Dumb” push notifications are virtually indistinguishable from spam. Dumb push stems from batch-and-blast mentalities that don’t take into account the user’s preferences, engagement, or habits. Dumb push leads to opt-outs and uninstalls. Nobody wants dumb push.

The only way to have smart push, at scale, is by utilizing machine learning. One way machine learning helps optimize mobile campaigns is by  enabling marketers to deliver personalized and relevant messages to users at times when they are most likely to want to receive them.

Optimizing message copy is a great example of how machine learning can do the heavy lifting in your marketing campaigns. A/B testing is vital, but the traditional process just doesn’t work for the Mobile Era. Traditional A/B testing requires a lot of time to start and stop testing cycles, and, by its nature, ensures that a significant portion of your base is receiving non-optimal messages at any given moment in time.

This issue snowballs as you create multiple campaigns because the marketer is always starting and stopping to tweak campaigns based on A/B results. Machine learning technology can automatically find the best-performing copy and tweak the distribution in real time, minimizing the amount of manual work and analysis that a marketer would normally have to perform. This is crucial in the Mobile Era, where data is collected from users on a 24/7 cycle, and every second saved is potential revenue earned.

Timing is everything when it comes to push notifications, and this is another aspect where machine learning can help to optimize campaigns. The problem lies in the fact that the “right time” for each individual can vary dramatically. A third-shift worker may want push notifications at 4 a.m., while others may immediately delete your app if you send messages before the sun comes up.

You can attempt to handle this problem by segmenting your users based on when they use your app, but this results in large groupings that don’t accurately capture user behavior, particularly as it changes and evolves over time. Machine learning is the only scalable way to ensure you can determine in real time the ideal window to message all of your users.

What’s Next

Along with personalization, timing, and location, marketers can further optimize their campaigns with deep linking and machine learning. Deep linking and dynamic deep linking greatly enhance the user experience by removing points of friction. Machine learning is important because it automates A/B testing and timing aspects, allowing the marketer to focus on areas where they can better add value, like message copy and brand strategy.

Optimizing your campaigns is important for ensuring messages resonate with users in ways that lead to virtuous actions, but these actions must be tied to a larger overall engagement strategy. In our next lesson, we’ll discuss how to develop engagement strategies that are tied to your acquisition channels.

Push 101: A Beginner's Guide to Push Notification MarketingPush 101: A FREE Beginner’s Guide

Master the basics and discover what push notifications can do for you and your app. Download our free guide today!