The retention rate for mobile apps – that old chestnut – is always on our minds. And so it should be! Cost per install (CPI) is only getting higher, and more importantly, the cost of a loyal, spending user can be an eye-watering multiple of CPI. If you don’t have the safety net of a smart and solid retention strategy in place, you may as well put a large amount of your acquisition budget straight into the shredder.

A common journey for an app user is something like: they discover the app, decide to download it, and then around 25% use the app only once before deleting it. Over the next 30 days, more and more new users fall away until you’re left with nothing but an empty bottle, a broken heart and lost users on your mind.

A sensible retention strategy is all about changing this familiar pattern, by segmenting users and targeting them with messages that can make the difference between lost and loyal customers. And remember, even if your retention numbers are only marginally improved, it can still make a massive impact on your business. In this area, movements of 2 or 3 percent can be the difference between success and failure. But how do you begin to implement a good retention strategy? Like most problems it helps to take a step back to widen your perspective.

How To Think About Retention

I’m sure most of us did something very similar the first time we used Google Earth: found the house we lived in, zoomed out so that the house became part of a bigger estate or area, which disappeared into a town or city, then into a country, a continent, and finally our planet as a whole in its full, colorful splendor.

It can help to start with that big picture when it comes to managing the retention challenge. That’s why Swrve’s User Lifecycle dashboard was designed to provide that perspective. The entire active user base of your app can be viewed from this all-encompassing distance. However, the active user base is then broken down into segments of a standard app lifecycle, bucketing users into six basic states of activity and inactivity. Taking this approach helps get a very quick picture of how your app is performing in certain key areas. It answers key questions such as “How many new users are progressing to an engaged state?” or “What percentage of lapsing users am I saving from churn?”. Those are benchmarks you need to be aware of.

As an additional benefit, segmenting users in this way enables you to target these groups with bespoke marketing campaigns relevant to their state. Different users in different states require different messages; it’s all about understanding what each state requires. Let’s take a look:

New Users

Or users who have only used the app once, which was within the last 7 days. Onboarding is, of course by far the most important thing to get right for new users. How you decide to structure your onboarding flow is another matter, but high performing in-app campaigns include welcome tours and optimized opt-in requests. It is worth considering delivering onboarding based on what channels users were acquired from. Depending on the channel, they may have different levels of understanding and sophistication. Similarly, you may wish to test whether it is appropriate to enforce early registration – or leave that for later.

Evaluating Users

Or users who have used the app 2 to 3 times and have been active within the last 7 days. Also a crucial phase, these sessions are more about subtly helping the user along and ensuring they are aware of the full scope of mobile functionality. The more value they are receiving from the app, the less likely they are to churn early. For example, have tip overlays point out features that certain users have not used yet, or remind them with an in-app message of the importance of opting in for push notifications and location tracking to get the most out of the app.


Or users who have used the app more than 3 times and have been active within the last 7 days. Whilst ‘engaged’ sounds like it’s time to slip off the loafers and start sipping a negroni, the reality is that your job is far from done. Do not take it for granted that an engaged user is going to stick around; the typical smartphone is a busy place with a host of alternative apps and services competing for attention. At this point in time, test push notification campaigns that maintain engagement (say, alerts about relevant media content) and use both push and in-app campaigns to drive users towards purchase – if that’s a relevant goal for your business. Purchases don’t just mean revenue at this stage – they also mean a greater chance of the user sticking around as their own commitment has increased and the ‘friction’ in tasks such as credit card entry has been overcome.

Bounced Users

Or user who have used the app once but now haven’t been active for 8 to 30 days. It is very easy to throw in the towel with bounced users; the ones that just didn’t seem interested from the get-go. But this isn’t good enough. Often all they need is a nudge back into the fold, and the reality is that they were not lost, just never activated. Entice them back via a push notification and a customized secondary onboarding to show the the real value of your app this time. Or perhaps they were simply distracted by a Snapchat notification (mobile is a medium of distraction after all) and then forgot to come back to your app. Send them a personalized push notification to remind them that you exist!


Or users who have used the app multiple times but now haven’t been active for 8 to 30 days. Ah-ha! We caught you sipping on that negroni and in the meantime users have lapsed. You’ve failed to convince them or remind them of the value of your app. It’s your duty to re-engage before they risk disappearing forever. Luckily you’re still in recent memory, so there’s a chance. Based on their (limited) activity with you to date, deliver smart, personalized push campaigns that focus on those interests and tasks – and A/B test different creative approaches!


Or users who have used the app one or more times but now haven’t been active for 31 to 60 days. Time is really starting to slip away, and previous campaigns have not worked. It might be worth being honest here – in most cases a user inactive for this length of time is probably trying to tell you something – and that thing might not be fixable with a simple marketing campaigns. However, in a multi-channel environment, where mobile and email data can be easily shared, it is worth sending disappearing users an email from the last chance saloon. Perhaps they are more likely to respond to that channel. To take another approach – and if it makes financial sense – special offers and discounts can be the option of last resort. Be careful in this area though – it is vitally important to ensure you don’t deliver financial incentives to users who were returning anyway. Track campaigns carefully and use analytics and split testing to ensure that’s not happening!