One of the biggest challenges as an indie app developer is getting your app discovered. My game Falcross has been doing well on the App Store for over 2 years, but download growth is flat. I’m on a mission to change that, and to share my progress with other developers.
Falcross is a puzzle game where you use logic to uncover hidden images. You get 3 puzzle “packs” for free and buy the rest via IAP ($5 unlocks everything). There are also earn-able/purchasable “tokens” which you can use to get hints, revive when you lose, and more.
The IAPs perform well, with content unlocks and tokens selling at about an equal rate. That’s great news, because it means I have value to trade for users’ shares and invites.
Falcross had no share or invite features 3 months ago. I’ve added some in the last few releases, and this is how it’s going so far:
1. Tweet to Revive
When you make too many mistakes on a puzzle, you lose and need to restart from scratch. This can be a 15-20 minute endeavor on larger puzzles. The game offers you a “revive” for 15-25 tokens (~$0.50), which erases your mistakes so you can keep solving where you left off.
As of 4.7, you can also tweet to revive once per day.
The game generates a nice twitter-optimized image of your progress for you to tweet:
— Micol re (@micol_re) November 17, 2014
So, are people using the feature? Let’s see:
The blue shows how many people saw the tweet preview and dismissed it, and the red shows how many people continued on to tweet.
Hmm. 10-20 successful tweets per day. Let’s see how that compares to paid revives using tokens:
See that little green sliver at the top? Yeah, me either. 10-20 tweets per day and ~400 daily revives means that only about 4% of revives came from a tweet. In other words, users would rather spend a bunch of tokens than send a tweet for free 96% of the time. (Note: This stat could be a bit deflated, as tweet to revive is only available once per day and token revive is always available)
Conclusion: 10-20 tweets per day isn’t necessarily awful, but I’m barely getting clicks to my bit.ly link from Twitter. That’s because most of the people who tweet to revive only have a few followers, and they do it every day. This channel needs work.
2. Invite Friends via Profile
The Invite Friends button is shown to logged in users, who account for about 6% of DAU (~250 logged in users per day). Here’s a daily graph of invites sent from this screen:
Not a lot of activity, but these invites are non-incentivized. Interestingly, Messages is the most popular way to share invites, followed by Twitter:
Conclusion: Not bad for a non-incentivized channel, but overall has generated <10 clicks to my bit.ly link in a month. Results are likely skewed by the inclusion of an iOS app banner on the shared invite page–I can’t track clicks on that. I want to replace the invite page with a Branch deep link which will handle following the inviter automatically without requiring the invitee to return to the invite page. This sharing option also needs to be expanded to anonymous users, who account for 94% of DAU.
3. Share to Unlock All
What if, after completing 50 puzzles, you could just share the game to your Facebook or Twitter feed and unlock every puzzle pack for free ($5 value)? This is absolutely the craziest viral strategy I’ve tried, and it works. Targeted at new users only, it’s the most effective sharing incentive I’ve found yet:
This channel generates about 10-20 shares per day, all from unique users. It’s also generated over 60 clicks so far. I don’t have referrer data for those clicks, but I believe that a lot of them come from Facebook, as I’ve heavily optimized the shared link for Open Graph:
So that means lots of people are sharing to Facebook, right? Wrong.
I used a UIActivityViewController to handle sharing here, with all activities except Facebook and Twitter excluded. Unfortunately, that doesn’t exclude iOS 8 extensions, so people have been “sharing” to Flipboard, Evernote, and Pinterest and still being rewarded.
Conclusion: This channel works really well, but it needs to be limited to Facebook. If users are getting a $5 IAP for free, they better earn it by sharing in a way that will generate clicks! I’m excited to see what happens once I enforce this.
Interestingly, this promo doesn’t seem to be impacting IAP sales much, but I’m still A/B testing with Leanplum to find the optimal time to show the offer.
All the charts above are from Localytics, my favorite analytics tool. It’s very helpful, but I also need to be able to track clicks and installs, both of which are external to Localytics.
I’ve decided that bit.ly link stats really suck. You can’t filter referrers by date range anymore (why remove this??), and the Twitter referrer data is a total mess.
My friend showed me Branch, a free service which claims to track referrals from share all the way to install. Organic referral tracking is the key to quantifying my viral growth efforts, so Branch is absolutely going into the next version of Falcross. I can’t wait to see what kind of data I uncover.