Pokémon GO's Impact on Smaller Developers

Pokémon GO has topped download and revenue charts since the day of its release and captured an unprecedented amount of media attention (even making the cover of Newsweek). It has also occupied an inordinate share of users’ device time. You might think this would greatly disrupt the mobile gaming market, but App Annie recently reported that “Pokémon GO has not had a sustained and meaningful impact on the daily revenue of other games on iOS and Google Play”. This has a lot of face validity for two main reasons: Pokémon Go has brought a lot of new players to mobile gaming (e.g. people who never played a mobile game before), and App Annie included revenue and audience estimates for mobile game powerhouses like Supercell and MachineZone. But the question remains: what was the impact on smaller publishers and games?

We all know that the mobile game industry has become a hit-driven business, and those hits can have a major impact on other developers, from the majors in the top 10 to the indies who are content with 10,000 monthly active users. In this post, I’ll report on the impact of Pokémon GO’s release for three of our customer games, and discuss whether and how you can inoculate your game against the next big hit.

The impact of a new game release on an existing game depends on several factors, including the strength and popularity of the newly released game, as well as whether its target audience overlaps with the incumbent game’s target audience.  Depending on those factors, we might  see changes in user engagement - both number of sessions and session length. We also might see fewer new users and lower retention of existing users. Finally, we probably expect lower revenues, though we might see an uptick in the revenue per user, since those who stick around are probably more intensely interested in our game.

To measure the impact of Pokémon GO’s release on the mobile game market, I analyzed data from three of our customers’ games: a shooter, a match-three action game, and a puzzle game. Each game was from a different publisher, and each game varied significantly from the others in terms of weekly active users. Though this is by no means a representative sample of all smaller developers and publishers, it should still provide some insight into the fortunes of games in the  long tail of the mobile game ecosystem in the aftermath of Pokémon GO’s release.

Approximately 500,000 new users appeared in these games within two weeks of July 6th, and I pulled the game activity data for all of them. I then split the users up into weekly cohorts based on their join date, spanning from two weeks before to two weeks after July 6th. I kept all user activity that occurred during the same week in which they joined, but discarded the rest. Thus within a given cohort, I had as much as seven days and as little one day worth of activity per user.

From these raw data, I calculated several common performance indicators for each game and cohort: A count of new users, the total sessions played, time per session, total revenue, and several more. Each of these is illustrated below. To preserve the anonymity of our customers, I am reporting the values in relative rather than absolute terms, using the “two weeks before release” cohort as a baseline.

There’s obviously quite a bit of variation. Arguably, the number of new users increases in the week of Pokémon GO’s release, while ARPU tends to drop. However, we only see consistent and statistically significant changes in the KPIs across the bottom row. Sessions and time per user fell 11% and 9%, respectively, after holding fairly constant in the two weeks prior to the release. The retention numbers are curious. We would expect to see a drop in retention for the user who joined the week before release - after all, they would be distracted by Pokémon GO, and thus be less likely to log in again in their second week. In fact, we don’t see a drop in retention for the cohorts before release, but rather in the cohorts after release. Why might this be? Perhaps players who started in the weeks before Pokémon GO’s release took the time to really get invested in these games. In contrast, those who joined after the release played fewer sessions and spent less time in game. Therefore, they are less likely to stay retained.

Overall, the effect of Pokémon GO on these particular games is hard to parse based only on these data. It is certainly difficult to tell whether revenue was positively or negatively affected by the release. That said, the  10% reduction in retention and engagement could have knock-on effects that last well beyond the month of July. And while these are not dire numbers, they are somewhat more stark than what App Annie reported for the entire Android ecosystem. It would appear that smaller developers and publishers were more vulnerable to Pokemon GO’s release than were the big guys in the top 10.

All of this raises an important question: Is there any way to inoculate a game against the effects of major new releases? Mobile game companies have two strategies available to them here. One concerns the scheduling of content updates, and the other concerns offers and notifications specifically designed to increase retention and monetization of the existing user-base.

Schedule Content Updates Wisely

Content updates are all about retaining existing users and pulling old users back into the game. They will have the strongest positive effect when these users don’t have other, new games to distract their attention away from your game. Therefore, it’s worthwhile to pay attention to upcoming major mobile releases and avoid scheduling content updates in the same time frame. How much time? Well, that depends on your game and your audience, but it's safe to say that just about every game company should have avoided updates around Pokémon GO’s release date. Developers of children’s games beware: Nintendo’s planned Q3 release of an Animal Crossing app will probably steal away your audience. Schedule your content update well before or well after, if you can.

Make Use of Special Offers

While releasing new content in the midst of a major new game release is a bad idea, efforts to retain players, increase conversion, and reward repeat purchases can pay dividends. On the theory that people have a budget of both money and time, the trick here is to tap into that before the new release drains it dry. That means splash screens, interstitials, special offers, and push notifications. Developers are often admonished to get new players to buy in early by offering “no-brainer” purchases, but in the time just before and after a major game release, this takes on a new level of urgency. Use offers to promote early buy-in, even if that means less revenue on first purchases than you’d like. Players who are invested are more likely to be retained - and to buy again later. Also, reminders of daily bonuses and other rewards will keep your game in front of player’s minds. So even if you usually avoid push notifications, you might consider using them in the weeks after a major game’s release. Reward the players who stay loyal, and give those who don't stay a reason to return soon.

Learn Something

The next time you’re in the midst of a major release, take the time to learn about the impact. First, go back and look at your data during the release of games like Pokémon GO or Clash Royale, and try to figure out what happened. Hypothesize ways you could perform better, then test those ideas during he next major release. Use what you learned to improve your response the release after that.

Be aware of the market

Implicit in learning something is being aware of coming changes. If your competitors just released new content, you should know that! If a major release is coming that will affect your audience, you should know that, too. Any number of gaming websites keep a calendar of upcoming releases. Bookmark that page and check back often. Also, do your opposition research. Be aware of the games that compete for your audience and check up on them on App Annie. When they release new content, you should respond accordingly.