Today sees the launch of the Swrve Predictive Suite, which brings the power of propensity analysis to the mobile marketing space.

If you’re asking what exactly ‘propensity analysis’ is, you probably don’t work in digital business. Most people who do understand just how important it is to know - or at least be able to guess - what users will do before time, rather than basing all marketing activity on what users have already done.

A few examples might clarify why that really matters. Consider the possibilities when you are able to identify:

  • Users more likely to stop using the service in the near future
  • Users more likely to make a first purchase or subscribe - if currently using a service for free
  • New users more likely to become VIPs in future

In each case a propensity analysis enables us to both identify these people and then communicate with them using the various marketing techniques and tools at our disposal. So for the three examples above, and in the mobile context, we might consider the following:

  • Deliver in-app messages highlighting features that a user about to churn has not yet discovered
  • Send relevant subscription offers to these users - potentially not offering a standard incentive and thus improving ROI
  • Deliver the best possible user experience, perhaps including random acts of kindness, to future VIPs to ensure they stay put

Of course further targeting can be added to those segments built on propensity, but it should be clear how powerful this ability is. And remember, there is a flip side. Consider those users who are NOT considered likely to make a first purchase. It is precisely those groups that we can then offer heavily discounted purchase options, safe in the knowledge that we are unlikely to cannibalise our future revenues.

How Does It Work?

One of the benefits of a platform like Swrve is the sheer insight it gives you into mobile users. We collect billions of events a day (in fact, for some customers we collect billions of events a day from a single app!) But that insight doesn’t just have to look backwards. By examining historical data we can determine - for each individual app - what different types of users looked like before key moments in their journey.

By creating these profiles, we are then able to look at current mobile users and identify those who are more likely to make a first purchase, churn - and so on. All automatically, and without any requirement on the part of a Swrve customer to manage that process. All they see is the segment of ‘Likely Purchasers’

So What?

Hopefully it is reasonably apparent that having these segments available to us can increase marketing effectiveness in a significant way. Even at the most basic level, we’ve seen push notification engagement rates increase by over 200% when they are sent at the right time for the user (in other words, when past history tells us that individual is most likely to use the app).

In the same way, predictive is set to move the metrics that matter right across the board. From engagement, to retention and (of course) revenue.

But it’s more than that again. Too often marketers think only of the benefits that accrue from a campaign, and forget the costs. But we should remember that in-app messages and push campaigns, almost by definition, distract the user - and that distraction can turn into irritation pretty rapidly. Only by being targeted and relevant will consumers come to love these campaigns.

In the past if we’ve wanted to drive subscription rates, for example, we’ve simply sent a message to everyone in order to make that happen. But that approach maximizes irritation. Imagine being able to reduce that irritation almost to nil whilst retaining all the benefits - because we are delivering campaigns solely to users who are ready for them?

That’s the promise of Swrve Predictive Suite. If you’re interested, drop us a line.