Our recently released mobile benchmarks report covered a lot of ground, comparing the mobile ecosystem by vertical, country, and OS.
But, to date, we didn’t dive into deep enough detail, and now we want to spend some time looking at how each vertical performed across the year. So, with this post we’re following up the mobile benchmarks with a look at Games. This will give you more data, and a better overview of how a single vertical changed as the year progressed.
Notes on the methodology
Below you’ll find a full year’s worth of data from 2017, covering retention, sessions and events triggered for Gaming apps on a per-user basis. You can use this set to compare against your own results, as well as spot seasonal trends as they change from one month to the next.
We didn’t delve into Game subcategories. While there are several of them, with all different kinds of performance, and events, we didn’t want to dilute our dataset too much. So subcategories including Action, Adventure, Arcade, Board, Card, Casino, Educational, Music, Puzzle, Racing, Role Playing, Simulation, Sports, Strategy, Trivia, and Word games aren’t individually featured here.
With that, let’s look at our first metric: retention rates.
Does retention change month to month?
A retention rate shows us the number of users that have stuck with an app over time. It’s useful for marketers because it tells us how well user interest is maintained. For apps that focus on lifetime value (LTV), retention rate is critical, but a lot of Games target the short term. Often, they are only interested in keeping users playing for a week, or even less.
In 2017, Gaming apps did two things: they had the second highest amount of retained users on day 1 and the steepest drop off of users by day 7 (compared to other verticals). However, by month, retention was practically identical, with up to 3% difference between months. Here’s what this looks like visualized:
Retention performance doesn’t change much for Games - some months perform a little better than others, but they don’t often overlap.
To illustrate our first point, we looked at retention in January 2017 and compared Games with two other high-performance verticals in order to see what the drop-off looks like:
It’s a much steeper curve, right? Bear in mind that out of the three, retention consistently falls; it doesn’t level out as with other verticals.
In a table, the drop-off looks like this:
|Day after install||Business||Drop off||Entertainment||Drop off||Games||Drop off|
What’s apparent is that drop-off on day one and day two are much larger - day four has the same loss of users for each vertical, but between the first and second day after install, 10% of users stop playing Games. In other months of 2017, we saw a 7% difference between day one and day two, and 2% losses per day as opposed to ~1% in other verticals.
That said, between day of install and day after, more users stick with their game than most other types of apps. In January, 59% of users had lapsed before their first full day of use. For Business apps, this loss of users was 76%.
What’s apparent is that new user acquisition for Games is paramount while the market remains competitive. Could retargeting be the difference? We’ve shown it’s effect, but many app marketers aren’t yet putting this strategy to good use quite yet.
How many times are Games played per day?
Sessions are one of the hardest datasets to translate into graph without introducing some bias.
Here’s a version where we’ve scaled the y-axis from zero. We usually see about 2.5 sessions per user, per day, so what does this look like?
Like retention, we can’t see too much variance on a wider scale. Can we spot nuances if we zoom in?
Close up, we can see that on day-after-install, the biggest difference between sessions is about .14, seen during the last two months of the year. We also see sessions dip about a week into use, though they rise after two weeks.
This is probably due to less interested users dropping off, leading to proportionally more-engaged users coming back, improving the number of sessions overall. Each of those bumps is also usually down to app marketers re-engaging users, who play for a day or so before disappearing again.
December’s behavior is different, receiving the lowest number of sessions on day one, but rising to second highest by day 30. This behavior could likely tie into events, so let’s see if that’s the case.
What’s Game engagement like?
Is it any surprise that December sees the most events triggered of any month of the year?
On average, users in December triggered 14.79 events per day. This includes level completions, upgrades, and more - and shows that in the last month of the year, users were more highly engaged. December events are at least double that of any other - and, in general, Q4 rises above the rest.
There could be a few reasons for this: utilization of a year’s ends worth of budget for promotion, new phones (or vouchers) as presents, holiday-season boredom - essentially, December is a unique period for Gaming apps, and a time heavily capitalized on by developers.
Apple itself reported on an upswing in activity in December, with over $890 million spent in the App Store between Christmas Eve, and the New Year. This coincided with the redesigned App Store, introduced with iOS 11 in September, which made Games much more accessible, and discoverable. The effect on events as a whole was somewhat profound:
As it stood, iOS saw much more action in terms of events triggered vs. Android in December 2017. However, Android also had much more events made in December compared to other months, at around two per day more compared to November 2017.
Whatever the reason, December is a special month for mobile, and one to be ready for in 2018.
For more data on Gaming, take a look at the Mobile Gaming Benchmarks report here, and discover how Gaming apps perform in the real world