Boyfriend Dungeon’s $272k Kickstarter Postmortem: Everyone Wants to Kiss Swords

A look at what worked, what didn’t, and what we learned.

Hey! Victoria (Kitfox’s community developer) here again.

Soooooooooooooooooooo. Boyfriend Dungeon’s Kickstarter did pretty well.

We’ve already talked about the previous work that went into making Boyfriend Dungeon a success and looked at the social stats from when we first announced, so now it’s time to look at what it all looks like now that it’s finished, look closer at some of the decisions we made, and where we go from here. Keep in mind survivorship bias is still a valid thing to look out for!

Expectations

Boyfriend Dungeon, by the time we launched, had already been worked on for a year. Strictly speaking, Boyfriend Dungeon could have been made without a Kickstarter at a bare minimum — five weapons, basic combat, a few dates — but that wasn’t what we dreamed for it, and it was proving to be an expensive project. It is, after all, two separate games mashed into one.

Boyfriend Dungeon had been rejected multiple times by publishers during its conception, and we were afraid its virality could be a one-off thing… we were wary of failure despite promising outlooks. So the Kickstarter’s goal was:

  • Community building and awareness. This was our primary goal.
  • Money to develop additional characters/content/polish.
  • Gauge the risk of continuing to develop this project — Was there still interest? Did it connect strongly enough with people?

We were cautiously optimistic about Boyfriend Dungeon because even though Moon Hunters had done fairly well, lots of people were very vocal about Kickstarter being “dead” and not being a viable thing anymore. But Boyfriend Dungeon had garnered a lot more attention than any of our game announcements previously, so most of us guessed we’d hit the $50k goal and maybe even double it by the end of the campaign.

However, no one even thought we’d be funded within 24 hours, let alone 6.5 hours. And then we quadrupled our goal.

aaaaaAAAAAAAaaAAAAHHHHHhhh????

The Campaign

We had adequately prepared for a successful campaign, but we had not adequately prepared for an incredibly speedy successful campaign. Mentally we had expected the normal Kickstarter campaign trajectory — peak at the beginning, slow down in the middle, and then peak again at the end.

But it never really slowed down — we averaged $4–6k per day in the middle of the campaign, and we’re grateful for that. Scrambling over too much money is better than scrambling over none.

Boyfriend Dungeon’s funding summary

One of our biggest headaches was actually dealing with the stretch goals. We had prepared them, but we ended up scrambling over it — Were they actually enticing? Which were more important? Were they too low/high? Were there enough? How fast do we want to reach them? How many can we commit to without going over scope? We decided to post a thank you, hint at our next goal, but overall not commit to anything until the next day after we had all gotten some sleep.

(Rest is good, everyone.)

Where’d They All Come From?

I hope you like NUMBERS!

Points of interest:

  • According to one of our Kickstarter contacts, the conversion rate for social (% of pledged column) is usually less than 1% in games. Our Twitter alone accounts for 17%. Dang!
  • Boyfriend Dungeon’s top referrals in comparison to our previous Kickstarter with Moon Hunters were more community-oriented, likely from our increased community efforts over the past 4 years.
  • Our bit.ly link was connected to our social pages, AKA likely these were even further direct contributions from people already in the Kitfox community.

Okay, so those are the people who actually pledged money. But not everyone is able to pledge money either out of interest, financial situations, or any other number of reasons. But eyes and word of mouth can be just as important as pledges, so here’s where our visitors came from.

Direct: Visitors from Kickstarter and people being directly linked to the page from someone just copying and pasting the URL from the Kickstarter page. Could also include links thrown around in Discords.

Social: Twitter accounted for most of these visits (about 60%), with Facebook, Reddit, and YouTube being the next biggest in that order.

Referral: Visitors coming from the Boyfriend Dungeon website and any news sites (Polygon, RPS, PC Gamer, Medium, strangely Tumblr, etc.)

Again, this could be an abnormal thing — Kitfox at the time of the launch already had about 10k followers, so it’s more likely that a lot of our users would come from social as opposed to a new studio or one less community focused.

We had quite a number of visitors each day as well, never below a thousand.

On Aug 23rd we were featured on the front page of the Kickstarter website, hence the spike there.

Social Growth

The Kitfox community really grew from this campaign, which was nice.

From the beginning of August to the end of the campaign on September, we got….

  • Twitter: 1986 new followers (10633 → 12619)
  • Facebook: 88 new followers (3063 →3151)*
  • Tumblr: 301 new followers (40 →301)*
  • Mailchimp: 335 new sign ups (6533 →6868)
  • Discord: 820 new members (566 →1386)

All accounts marked with a * were not linked to the Kickstarter page. As a small explanation, I’m not fond of Facebook’s algorithms when it comes to posts, and I didn’t have much time to commit to Tumblr, hence them being left out.

What Worked?

  • “Limited Time” Early Bird: We decided to use the “limited in time” option for the two early bird tiers (one for just the game, and one for the box set) instead of “limited in quantities” — we wanted people to pledge as quickly as possible. People are willing to pledge to a project they think is likely to succeed and/or has already succeeded thanks to the snowball effect, and putting a time limit increases the likelihood of Day 1 pledges. It worked — overall we got 2333 Early Bird pledges. We figured the risk of having cheaper pledges would be mitigated by the amount we’d get, and it seemed to work out. People can always adjust to a higher one in the end anyway! It probably helped us make our 6.5 hour funding record too, which made it even more newsworthy.
  • An animated transformation sequence: Particle Beam did a fantastic job at making an eye-catching moment for people to latch onto.
  • Hidden stretch goals: Revealing them one by one gave us time to think about the costs, their scope, last minute changes, and gave us something newsworthy to announce everytime they were revealed.
  • Fun stretch goals: Not everything needs to be about virality, it’s worthwhile to acknowledge the community that got you there. Hair dyeing and bubble tea tasting weren’t huge funding pulls, but the community and super fans got to chat and create some inside jokes with us.
  • Bladesonas were a cute idea and kept the community engaged and interested during times we were between goals.
  • Previous Kickstarter experience, shipped games, and studio legitimacy were a HUGE plus for us. We had a bunch of people who explicitly told us they backed the game because they knew we would treat Boyfriend Dungeon as a serious project, and not waste its premise as a gimmick. Plus people (even ones who hadn’t played our games) would point out our previous repertoire of games to anyone who expressed doubts about our abilities or how robust/fun the game would be.
  • Our press emails were about our new trailer and two new dateable characters, and not about the Kickstarter itself. Even more than before, Kickstarters are not newsworthy items by themselves. Having something else prepared and a hook for press to latch onto helps!
  • A Backer-only Discord. It’s only been a few days since we launched the secret Backer-only Discord channel, but so far there has been nonstop chatter and there are already 300 people in it. Granted most of the discussion is off topic, but it’s made our community feel much more intimate and dedicated. However adding people to this would’ve been a nightmare if the lovely team from Temtem didn’t provide information about their bot to us.
  • Newsletters, but they were love letters. This was a cute feature we had for Boyfriend Dungeon and a lot of people latched onto the idea. Not sure how much more it increased sign ups, but we definitely had people talking about them and probably got more engaged fans from it.

What Didn’t Work?

  • Revealing stretch goal costs too early. When we first announced our stretch goals they were secret, but we DID announce their cost upfront. We should’ve kept the costs hidden to account for any changes that would come up later.
  • Trying to price in USD. We thought pricing it from a USD standpoint would be better since most of our backers came from America, but it ended up making our numbers look all funky (e.g. a $65,579 funding goal) and due to currency fluctuations, sometimes our CAD goal would be reached but the USD one would be lagging behind. It made for some confusing moments where backers weren’t sure if a stretch goal was funded or not.
  • Twitter did.. a thing?? About 3/4 of the way through our campaign, an odd thing happened with our original Kickstarter announcement tweet (a big one I pinned with the most engagement and impressions.) It just… disappeared? For other people, but not on our end? I’m not sure how much that affected future engagement, since now people couldn’t see it and it appeared broken in retweets. (CRYING.) It’s extremely odd, since I can STILL see it from the Kitfox Games account, but when I go on my personal account or link to it, it only brings up an error page. Spooooky.

What Kinda Maybe Did Something?

  • Did having a “dev interview” portion of the Kickstarter video matter? It seems the talking portion of Kickstarter videos is going out of fashion, and while most people probably tuned out by the time the exciting trailer portion was over, it might’ve helped other people connect with us as humans in general.
  • How much did PAX West help with pledges? We didn’t see a HUGE spike from going to PAX West (Aug 31 — Sep 3), though we did meet a bunch of passionate fans and members there. Plus we had the most press coverage than we ever have before, which probably helped our visibility a few days after the dates of PAX or after the campaign was over, depending on when articles were published.
  • No add-ons. We wanted this Kickstarter to be as low-maintenance as possible so we can focus on the game and prevent getting too distracted and delayed… while also letting us make some cool pillow-cases and box sets mwheheheh. As a team of 8 (and only 4.5 of those 8 working on Boyfriend Dungeon), we honestly don’t have the bandwidth to handle micro-managing pledge tweaks and ensure the best service is provided, even if it meant we’d possibly get higher pledges. So it’s a give and take, but we’re happy to be working with Indiebox (who worked on the Moon Hunters Indiebox Collector’s Edition) for product design and logistics handling. Meanwhile, we chose itch.io for its robust system to handle digital fulfillment. We’ve worked with both of these places in the past and they’ve both been excellent!
  • No console goals. We might’ve made more money if we had console stretch goals, but we wanted to focus on the base game and not repeat any mistakes we made with the Moon Hunters Kickstarter and its console promises.
  • Having the “Now on Kickstarter” appear in the trailer. Tanya and I had a small debate about this and whether or not it should show up at all. Should the trailer be evergreen and risk not having the clear CTA, or should it have the clear CTA but look outdated after a month?
  • No “$1” pledge tier. A $1 tier could have caused an anchoring effect for those that saw it first, making pledges smaller or making them less willing to spend money at a higher tier. You can always pledge at $1 if you really wanted anyway. We still got 62 pledges with no reward, and a bunch of those were $1.

Final Thoughts

One thing I want to be clear about is not to get hung up on just the numbers. They’re a way to gauge success, but I don’t believe they should be all and end all. A failed Kickstarter doesn’t necessarily mean a failed game.

Not everything we did during the campaign was for virality or to get more funds. Some of it was to solidify our community and let them get to know us a little better. Numbers can mean nothing if you don’t have positive word of mouth or an engaged community to do anything about it. Boyfriend Dungeon could’ve very well been a viral hit and then left people immediately uninterested if it were not done genuinely or with no support. I’m glad that didn’t happen.

And now?

It’s time to make a game. ❤