Design to Share: How Viral Sharing Works in Blackbox

A note from Ryan: If you’re reading this chances are you’re a part of a larger iOS or design community I’ve been quietly learning from for quite some time. I owe you a great deal and I think it’s my turn to start giving back and sharing more. Here is the first of a few drafts I’ve been holding onto far too long.

I added sharing mechanics to Blackbox as an afterthought—but I’ve been designing for sharing since day one. To this day, these design decisions and mechanics keep Blackbox growing and reduce the need for marketing.

Below is a breakdown of how sharing in Blackbox works from a mechanical level, as well as some of the design considerations I made along the way. Many of these observations may appear to apply only to games, but I firmly believe they hold merit for much more.

Never Played Blackbox?

Blackbox is an iOS puzzle game for practicing creative thinking by solving puzzles that generally don’t involve touching the screen; it’s a growing collection of 70+ puzzles that are solved with things, places, timing, careful thoughts, and other outside the box thinking. It’s a hard game to explain and hard to get out of your head once you start. It’s an Apple Design Award winner and is free so why not give it a try?

Individual challenges are represented by “lights” that turn on and stay on once solved. Each light/challenge is unique and has its own hint if a player gets really stuck. These hints are revealed with hint credits which are provided to new players, can be bought, or earned by referring new players.

Blackbox’s Branch links are setup to reward players once their referred friends install the game.

Going to Africa

Two years ago, when I was an aspiring indie developer, I knew as much about the state of the App Store as one can know from the outside looking in. I knew that without a reputable name, paying for users, or a marketing team to foster success, I’d need big ripples of attention/downloads — and something to sustain the resulting waves in order to survive long term. My strategy would be to make Blackbox inherently worth sharing without taking scuzzy shortcuts to get there.

One of the first inspirations for Blackbox was Inception: a promotional app for the movie in which you unlock soundscapes (dynamically blended using the sounds around you) by doing ridiculous things like opening the app in a rainstorm, late at night, or in Africa (the continent). “Who does that?!” my friends and I would bemoan, talking about the app and inevitably explaining it to one more person who was listening in, intrigued and confused.

The experiences in Blackbox were designed with this sort of conversation in mind, but the social degree to which people played went far beyond what I anticipated. Students set up dedicated group chats at their school, coworkers created Slack channels to share their frustrations, and couples got into loving tiffs over who solved what first. Each puzzle solved was a journey in itself — journeys that left behind stories of summiting mountains in the rain, families yelling and singing, and countless other “ah ha!” moments. What Blackbox lacked in replayability or casual play it made up for as a conversational centerpiece. Blackbox was inherently something worth talking about.

Breaking Down Barriers

I knew the app would be hard for players to explain to friends without spoiling, so I tried to eliminate the barrier to download. I made the game free, with a path for pro players to pay to unlock an additional chunk of challenges or reveal hints. My conservative monetization strategy might have been a naive one, but I’ve never regretted the decision to make Blackbox free. It’s made it that much easier to share.

My conservative monetization strategy might have been a naive one, but I’ve never regretted the decision to make Blackbox free.

The random arrangement of levels on “the grid” (see above photo) is designed to encourage players to explore and chase after curiosity rather than follow a linear progression. This also had the unintended side-effect of introducing friction when comparing scores.

When players want to compare scores they don’t rattle off the number of challenges they’ve completed, they post screenshots of their grids or pull out their devices to compare in person. Online, the screenshots add colorful intrigue to posts and make people ask: “Wait! How did you get that one??”. Blackbox’s dozen or so scattered meta puzzles only fuel that inquisition, as they’ll only display through obscure methods (have you seen Blackbox’s website on mobile?) that players like to share with (or withhold from) friends.

Telling a friend about Blackbox is easy, but to get someone to download it is a whole different level of effort. Blackbox offers plenty of opportunities to share with a link, but most people already in conversation are unlikely to go into the app to “properly” share it …at least not without incentive.

In the early days, Venmo paid out real money to invite people — something like $1–5 per person—and I’m sure many of us remember the days of fighting to refer a friend to Uber or Lyft (“that way you can get us all home tonight for free!”). We may have loved these apps already, but the extra incentive made it worth hopping through an extra hoop or two to seal the deal. I found a way to do this for free with Branch, and within a day added the ability to reward players for referring new players with hint credits (an item I could afford to trade). While the fraction of new players each day who come from referral links is a small one, the network effects have been huge.

The Numbers

60–70% of Blackbox shares and resulting viral installs happen via personal/group communication (chiefly Messages and WhatsApp according to Branch’s analytics). While it makes sense that Blackbox is more personally shareable rather than broadcast sharable, I also believe the attention I gave to placeholder text has some impact. Poorly customized placeholder text can be a barrier to sharing something, and few apps go out of their way to customize placeholder text for what’s being shared or for the platform it’s being shared to. Few of us would write “I scored 42 in Grouchy Penguins! Can you beat my score??” so why offer that as the default? And would we write the same thing if we were new to a game and our score was low?

60–70% of Blackbox shares and resulting viral installs happen via personal/group communication

Actual players are the greatest source of inspiration for Blackbox’s ever changing placeholders, leading to some amazing ones I would have never dreamed up. Some samples are the uplifting, “I’ve lost my life to @BlackboxPuzzles” and the even more cheerful, “This game will ruin your life”. Some of the placeholders even include emojis that become increasingly boastful the higher the score is.

3.5–10% of daily active players choose to share Blackbox via in-game interfaces (which are likely to include a referral link); those referral links convert into tracked installs roughly 13% of the time. Over a four month period that lead to 110,600 viral installs via Branch links, or 4.3% of new players in that period.

That might not seem like a poster child of virality, but is made more significant when we consider the exposure created through all that sharing, and the subsequent untracked installs. Even more significant is that players who install via Branch links are ten times more likely to purchase something in app than those who don’t. Further refinement and personalization of the referral system could take this even further.

Fun facts

• The biggest Blackbox influencer has directly referred 469 players who have subsequently referred 99 new players… and so it goes.
• 336 players have referred 10+ players.
• In total 112,454 influencers have referred 164,884 new players.
• At this time of writing Blackbox has had just shy of 4,000,000 downloads.

Spoilers Ahead (actually though)

The numbers weren’t always this good, but the inclusion of a recently added challenge helped. For a long time, I thought it’d be fun to have a challenge that presented the player with a QR code and scanner. The challenge could be solved by scanning the QR code of someone that player had referred. Before referring someone, the scanner would be disabled, and once scanned, the QR code would be consumed and crumble apart. The mechanic was intriguing since it involved a level of cooperation the game had never had, but it always felt a little ham-fisted.

A few months ago, I had a better idea: a challenge unlocked after sharing where players would be assigned one of three QR colors (pseudorandomly based on a player ID) with the implied goal of collecting all three colors. Clever players figured out how to scan their own QR codes to get one of the colors, but were then left to the task of finding the rest. While understandably not everyone cared to refer friends for hint credits, many enthusiastically got several to a dozen new friends to start playing to see what color QR code their Kinder Egg contained. It also flooded the Blackbox subreddit and twitter mentions with a deluge of daily QR code swaps.

The three different possible QR code colorsseen in the “Buddy Scan” challenge.

Slight of Hand

In an App Store chock-full of money-extracting skinner boxes thinly veiled as games commanded by an industry poised to overtake gambling, it’s exceedingly impossible for indies to succeed on merit alone. I believe many of the ideas I mention are the fuzzy fine line, the silent differentiators, between those who succeed and those who do not. I’ve always wanted people to share Blackbox on their own volition but elected to take no shortcuts in achieving that. I hope this serves to prove that there are always clever solutions to succeed without compromise.

I don’t have an email newsletter but if you’re interested in content like this please follow me on Medium and Twitter. More to come…

This post was originally published by Ryan McLeod on Medium. Here’s the link!