Illustration of Derek Powazek by Adam Ellis

Alphas Are Experiments

cute-fightCute-Fight is currently in private alpha. That means the whole site is locked behind a password, and you have to request an invitation to join. People often think that a private alpha is just cover for an unfinished product, but that’s wrong. A private alpha is an experiment with a specific goal. And like any good experiment, it starts with a hypothesis, and failure is a valid result.

The main question in the Cute-Fight alpha was: Will anyone do this? The answer has been a resounding “YES!” Our small pack of early members has already created hundreds of fighters, fought thousands of fights, and cast over 30,000 votes.

But that “yes” had caveats. We included a form at the bottom of every page that said “Give Us Feedback.” Anything put in it was emailed to the team. And we’re learning a ton from our early Cute-Fight members (who are all smart people with adorable pets, by the way).

So I decided to write down three interesting things we learned from them and what we changed in response. Later, all of this will look obvious, as everything does in hindsight, but right now it’s an exciting series of discoveries.

“How do I start a fight?”

Discovery: A lot of early members had no idea how to start a fight.

Context: The site is set up for Cute-Fights, but we also wanted people to be able to opt-out if they didn’t want to fight. And you can’t start a fight unless you had a pet listed yourself. And even then, the pet had to meet certain criteria (like having enough photos). As a result of all of this complexity, you could start a fight from some fighter profiles and not others. If you were just looking to start a fight, it was difficult and frustrating.

Tweaks: First we added some help text to the fighter profiles and upload system to explain that a pet had to have three photos in order to fight. We also took new members right from fighter creation into photo upload, which helped ensure most fighters had the minimum three photos. But none of that solved the core problem: people still didn’t get how to start a fight.

Solution: We created a single fight creation process to help people start fights. All they had to do was click the “Start a Fight” button, which we placed in the sitewide header. If they didn’t have a fighter, they’d be prompted to create one. If that fighter didn’t have enough photos, they’d be asked to upload some more. If they had a fighter with photos, they’d be shown a random selection of other fighters they could challenge. Once they selected one, they went into the normal challenge process.

Result: Overnight, all the “How do I start a fight?” questions stopped and fight creation went through the roof.

Unintended Side-Effect: The rate of new fight creation spiked. Some voters complained about seeing the same fighters in too many fights. To fix this, we just implemented a throttle on new fight creation that hopefully very few people will ever bump into. (When a fighter is in three concurrent fights, if the manager tries to start a fourth, they’re told that the fighter is tired and needs to rest.)

“Losing makes me sad.”

Discovery: Even though Cute-Fight is a silly game, and we’ve designed the site to reinforce the silliness, some early members let us know that losing made them feel sad.

Context: Games need to have stakes, but our goal was to make losing as fun as possible. That’s why we have badges for the fighter that didn’t win and other random rewards. Still, we had a problem.

Research: I contacted members who reported feeling bad and asked them some questions. Was it the fact of losing that made them feel bad, or something more specific? Where did they first have that sad feeling? (Side note: being comfortable talking about feelings is pretty helpful in startup life. Hooray for therapy!) The members reported that it wasn’t the losing that hurt, it was being reminded by the loss in big, orange numbers on their fighter profile page. They were already feeling ownership of the page (good thing), but having proof of losing on the page was a bumming them out.

Tweak: As an experiment, we removed the win/loss numbers from the fighter profile pages entirely. We wanted to see if anyone noticed. Only one person asked where they went. No one complained. And no new reports of “this makes me sad” have come in since.

This is only a temporary solution. Cute-Fight is a game, and games have scores. But we need to find a way to reward players for winning without making people feel bad for losing.

Aside: This is why so many games have unlocking achievement reward mechanisms. It rewards play with random reinforcement (more powerful than consistent reward), but does not punish poor performance publicly.

“Where are my friends?”

Discovery: Cute-Fight is a social game, and social games are more fun with your friends. But, in the beginning, we had no way for anyone to find anyone else. This was no big deal when there was one page of fighters, but when it became 20 pages, people started to ask: Where are my friends?

Context: The long-term plan has always been to have copious connection to social networks, with importing of your social graph and exporting of fight information. But because the site was locked behind a password during alpha, we removed that functionality.

Side Note: It amazed us how much people talked about Cute-Fight on Twitter and Facebook, complete with links, even though the site was inaccessible to most people (and we warned them in email this was the case). This is just more proof that you don’t need a pile of “SHARE THIS” buttons to get people to talk about your thing – you just need a thing worth talking about.

Research: I spoke with members making this request, and they were pretty evenly split. Some wanted to log in with Facebook or Twitter credentials and see who in their network was playing, some didn’t care about that. But everybody wanted to be able to search the site for people or pets they knew.

Solution: We’d had search on the To Do List for a while, but it was a fair way down. We’d thought it wouldn’t be important until we had a ton of members, but our members told us they needed it now. So we’ve moved it to the top of the list and hope to launch it soon. It’ll be simple and won’t solve the whole problem, but it’ll help. Full Facebook/Twitter integration is still on the list, but not until public beta.

Those are just three examples of changes we’ve made in response to feedback. There are lots more (like the player cards, revisions to the voting and cheering mechanism, and the slow load-in of photos). Like I said, the best part of skating to the puck is that, when it works, your community will tell you where they’re going. Your job is just to keep up.

Thanks to our early members for being part of this wild experiment. There’s more – much more – to come. Wanna play?

ps – We’re also looking for angels.


← Back to Home

Hi, I’m Derek. I live in San Francisco and make awesome community-centric web stuff. I sometimes post things to Flickr and Twitter. I’m mostly harmless. More.





This site is powered by WordPress and expertly hosted by Media Temple.