Welcome back to our Kickstarter retrospective on Sunless Sea. Last post had us just about to begin the campaign, and me using the crudest guesstimate in the world about our final numbers. Here's how we got to the finish line, and a few things that helped and hindered us along the way. There are fewer pictures of ships and more charts in this one, but there is at least a badass golem.

Kickstarter Eve

We announced the date and time we were going to launch, well ahead of time, so the community were stoked to get going. Let me skip back slightly and share one more practical tip here.

Once your Kickstarter's ready to launch, you can share the preview link with whoever you like. We shared this with our earlycore community types - the people who'd already given us feedback. Once again, we were very glad we did! There's always something you missed that someone else will spot. But we weren't aware that once the campaign goes live, the preview link will point to the actual campaign page. This meant that when we stealth-launched the campaign an hour ahead of the official launch, just in case something had gone wrong or we got delayed, anyone with the preview link could find the campaign, even though we hadn't published the final URL. This meant we went to our forums to announce it and found a half-dozen people talking about us already, which left us scrambling slightly to catch up. Not a big issue! and it was exciting to see we already had backers. But it would have been embarrassing if we'd stealth-launched the night before.

Getting the word out

Getting press for Kickstarters is hard. We talked to other indie devs running KS campaigns, a couple of who said: oh thank God! I thought we must be doing something wrong, it's not just us then? A Kickstarter campaign hasn't been a novelty to the games press for a long time, and to get publicity you need to be at least one of (a) lucky (b) unusual (c) really big. Some outlets even told us they have a KS embargo. So manage your expectations.

And wherever you can, treat journalists like human beings, not like email addresses on a list. We got some really nice day zero press from Adam Smith at Rock Paper Shotgun and Cass Khaw at USGamer - both of whom we'd talked to before, both of whom had done interviews with us before, both of whom we were as honest and open with as possible. Treasure your press contacts, talk to them. Aside from anything else, they're people too. Don't just throw press releases at a contact@ email and hope for the best. Find out who might be interested in writing about your project, get hold of their professional email address (this is often as simple as asking them nicely on Twitter) and write to them proper. For more general coverage, it's a good idea to get your press release on a site like http://gamespress.com/ Many smaller sites will use copy from release sites unedited, so make sure your release is clear and attractive. There's a lot more useful information about talking to the press here: http://www.gamasutra.com/view/news/198381/

An unsuspected reef!

On our previous KS campaign, we offered more physical rewards, and damn it was a lot of work fulfilling them. When you're used to doing everything in cyberspace, the reality of printing, posting, chasing up missing packages, eats up time. So this time round, we tried to keep our rewards digital where we could. This made sense to us: what Failbetter mostly do is story content, and our community is very used to some content being locked unless you pay for it. We had three tiers of content-only rewards, and we adjusted the pricing on those in response to user feedback.

In week two of the Kickstarter, we published an update about some of this higher-tier content, with some new and rather tempting concepts. We thought it was a nice way to do a bit of gentle up-selling and to share some of the exclusive art. We were a little bit worried about the Pirate-Poet, our 'gorgeous renegade poetry-tattooed non-traditionally-gendered pirate death-golem'. We were concerned that the first female-presenting Clay Man in the game was going to be depicted with her top off (to show those poetry tattoos), when all the male-presenting ones had been fully clothed. Paul and I went back and forth until we were confident we had an attractive, distinctive, non-objectifying image. We watched the comments to be sure that we'd taken the right approach.

So we were completely blind-sided when we immediately ran into a problem that had nothing to do with gender. The first comment called our rewards strategy 'shitty', the second pulled their pledge, and the third agreed with the first two.

After that, just as we were wondering if (a) the Internet had gone mad or (b) we had colossally misjudged our approach, other backers started to respond and to say, hang on, this is just a bit of extra content, and it's how Fallen London works, and anyway it was there in the rewards list from the beginning, why is it suddenly a problem? We let out a sigh of relief and responded politely. The comments thread got lively but stayed largely civil. Most of the commenters agreed that it was fine, and others were won over by our responses. But it was really the only sour note in a cheerful, successful campaign. And it could have been much worse. So what went wrong?

We've always been generous with content. The approach in Sunless Sea, as it is with Fallen London, is to make nearly all the game available to everyone, and charge a premium for something a little bit special. Our existing audience knew this and had seen us do it for years. But the audience who'd never heard of us before they came to the KS just saw a game company charging for 'day zero DLC' - slicing out a key part of the game and charging extra for it.

I guess the lesson here is 'assume nothing about your audience'. Most people don't know or care who Failbetter are, and we can't reasonably expect them to. If we'd added a para at the top of the update making it clear that the premium rewards were a fraction of the content - if we had added more concept art that was about the general as well as the premium content, so people didn't feel excluded - maybe that would have helped. We got more cautious from there on in.

The total is not a high score

If your KS is going well, the number will keep going up. Up and up and up! It's your validation, it means you can pay your team for another day, another week, another month! BUT DON'T FOCUS ON IT TOO HARD.

If you are a small or medium-small team making a game, time is your most important resource. You may need to spend money on software licenses or sound design, but the KS funds are probably mostly to keep you eating while you make the game. This means that if your overheads (team wages, rent, servers, everything) are (eg) £300 a day and you all spend a day working on something and you increase your funding total by £300 over what you'd have got otherwise, then you are worse off than if you had done nothing. You haven't stood still. You've lost a day from your delivery date, you've got more backers who will take more admin time to respond to, and odds are you've picked up some expenses for pledge rewards too.

It was really, really great that we got over-funded. It reduces our risk, and the 100K total made us more newsworthy. But we held off the temptation to add new reward levels or interstitial stretch goals (apart from one very unusual one I'll talk about in a bit). Maybe we could have shaken the tree a bit harder and attracted another 10K in funding by spending more time on concept art or demo videos up front. But I think we struck a good balance.

US vs UK Kickstarters

On the day we launched, Thomas Bidaux of Ico Partners published a blog post on data from US vs UK KS funding. I'm very grateful that Thomas waited until then, because if I'd read it before we started I would have been in a state of even more extreme anxiety than I was.

The short version of Thomas' post (and go and read all his posts, it's good analysis) is that UK Kickstarters do much worse. UK Kickstarter projects don't have access to the nicely frictionless Amazon payments system; plus EU backers are often used to paying for things in dollars, and US backers often aren't used to paying for things in pounds or euros. 2/3 of Failbetter's audience is US, so this probably hurt us.

We could have used a US-side contact to run the project. We'd done it before, but our previous contact wasn't available this time, and anyway there are disadvantages to using someone else's KS account - it means your campaign history ends up spread, potentially, across multiple accounts, it means people will wonder why the company name doesn't match the account name, the whole thing just looks slightly more sketchy. Given that we got funded, it's tempting to say we made the right decision.

But it also meant that when the US Government shutdown hit, two days before the end of the Kickstarter, it hurt us, because the dollar suddenly got weaker - US backers had to pay slightly more, and they were probably that bit warier about pledging. Currency fluctuations! They're not just for enormous transnational financial institutions!

Ports of origin

Data? Since you asked so nicely. Here's where our actual backers actually came from. (Well, the top 25 referrers that KS tells us about.)

Referrer	Type	# of Pledges	% of Pledged	Pledged
Direct traffic (no referrer information)	External	965	27.86%	£28,086.21
Video Games (Discover)	Kickstarter	587	9.28%	£9,356
fallenlondon.storynexus.com	External	471	12.16%	£12,254.53
rockpapershotgun.com	External	283	5.05%	£5,091
Search	Kickstarter	274	10.78%	£10,871
Twitter	External	202	4.39%	£4,425
sunlessseagame.com	External	194	4.54%	£4,577.15
Kickstarter user profiles	Kickstarter	124	2.72%	£2,741
Facebook	External	111	2.35%	£2,371
A project’s backer confirmation page	Kickstarter	106	1.42%	£1,429
reddit.com	External	91	1.31%	£1,322
48-hour reminder email	Kickstarter	84	1.19%	£1,199
google.com	External	70	1.51%	£1,525
Staff Picks (Discover)	Kickstarter	45	0.91%	£914
Home location	Kickstarter	44	0.75%	£751
Friend backing email	Kickstarter	42	0.65%	£657
mail.yahoo.com	External	41	0.95%	£961.50
google.co.uk	External	39	1.22%	£1,232
pcgamer.com	External	37	0.63%	£635
Popular (Discover)	Kickstarter	36	2.55%	£2,571
forums.somethingawful.com	External	30	0.60%	£600
quartertothree.com	External	30	0.48%	£485
Starred	Kickstarter	27	0.34%	£346
Embedded widget	External	21	0.58%	£586
jayisgames.com	External	20	0.31%	£315

So...

KS is a discovery as well as a funding platform. Almost a third of our pledges came through KS. Take that number with a pinch of salt - if someone read about us elsewhere, came to KS later and looked for us, they'd show up under 'Discover'. But no doubt KS helped raise our profile, and that plays into the other great thing about Kickstarter - it means that when we release Sunless Sea next year, we've already managed to publicise it.
Publicity on a major gaming blog makes a big difference.
That sunlessseagame.com promotional site of ours was well worth the effort - especially since we're not junking it now! We'll keep it around as long as we are building or selling the game.
And, once again, our existing community was huge for us. A big thanks to them. We think that around 30-40% of our final pledge total came from FL players in one form or another - it's hard to tell exactly - but the most important thing was that it's the community that got the word out.

A last-minute sea-serpent

You know those LAST HOURS stickers that KS creators put on their campaign images? It turns out Kickstarter don't like those.

"This is a message from Kickstarter Support. We're reaching out because we noticed that you recently updated your project image, and that the new image includes a 48-hours left banner. As a result of the change, our editorial team has removed your project from the Staff Picks section..."

Eeek! We fell off Staff Picks with 48 hours to go, just when the funding was picking up for the last-minute bump! We changed the image and they put us back, but we lost five hours of Staff Picks. Did that lose us £5? £500? £5000? We'll never know.

The reasoning, in case you were wondering, is that the image ends up elsewhere on the web with 48 HOURS LEFT on it, which means that people click on it six months later and get annoyed. You also can't change the image once the KS is complete, so it'll be frozen forever with FINAL HOURS on its image like a beetle in amber, leaving KS backers clicking and tutting when they realise they missed it. Fair enough.

Steaming home

Remember this?

"The day before the Kickstarter launch, I did a basic calculation based on our promotional site email list, on the very fudgy assumption that some of those 3000 people wouldn’t pledge, but about the same number of newcomers would pledge. The average Silver Tree pledge had been about £28. So 3000 x £28 = £84,000."

The first and last 48 hours of a Kickstarter campaign are the ones where you get most pledges - the first 48 when you make a splash, the last 48 when backers go, wups, nearly missed that, quick now! I'd been fiddling obsessively with a spreadsheet over the month of the campaign, and had estimated the last-days bump based on previous experience and similar KSs. The final total looked like it would be very close to that £84K early estimate.

Then we suddenly made £16K more than I was expecting.

So how did that happen? Here are some possible factors for our last-minute bump. Besides our community being god-damned heroes, which was significant but not unexpected.

We sent a mailout to our Fallen London mailing list 72 hours before the end. I'm embarrassed to say that this wasn't quite as brilliant a piece of planning as it seems in hindsight. We had meant to do it the previous week, but we'd been so busy we'd forgotten. So, yeah, mailing your base 72 hours before the end might be a good idea.
We did a Reddit AMA. http://bit.ly/1g2XfG7 This was surprisingly time-consuming, actually. I wish we'd focused more on Reddit earlier on. They're a clever crowd with real skill for sifting out gold from crud. I think I'd just assumed we wouldn't get past the surface froth of cat gifs, but the subreddits system helps a lot.
Early on, Liam and I both committed to getting tattoos if we passed 100K.

...yeah, okay, that last bit was an exception to the 'no extra stretch goals' rule that we set ourselves. But it won't take up company time, and although, honestly, I'd rather have ended with the total at a nice safe 99,990, I don't begrudge the idea of memorialising an experience like this. I thought 100K was going to be close enough to be tempting but far enough to be out of reach: turns out I was wrong. :-)

In case you were wondering, Liam's going for an anchor-and-tentacle arm design. I'm thinking calf, which is safely concealable and suitable for a runner. Someone's suggested a tentacle reaching up with a telescope, which I rather like. In the end, it's always all about the tentacles.

The final total

The last thing I want to say is not new information, but it bears repeating. We raised £100,800. After Kickstarter and credit card fees (which are predictable in the planning phase) and dropped pledges (which are less predictable), the final total is £89,616. Plus we need to earmark a couple of thousand for physical rewards. Plus, depending on how well we do this and next year, we will be liable for tax on it.

All told, setting up and running the campaign, plus building Sunless Sea, plus building the expansion pack we're now committed to, will probably be about 40 person-months of work. That's £2,190 a person a month. The UK average wage works out at £2,200 a month. And we've got software licenses and rent and other expenses to pay here too. We're not exactly drowning in cash.

That's fine! We always planned to fill in the gaps with income from Fallen London. And at the end of it we hope we'll have a game that we can sell. But the Kickstarter campaign is the beginning, not the end, of the journey. If you're thinking about running a campaign, you are never going to get more money than you can spend. Not many projects run under budget. Our responsibility to ourselves and our backers is to spend that money carefully.

So I suppose we'd better get on with it.

Thanks all for your enthusiasm and your support! We're heads down now until May - but we'll be posting updates about ongoing development to the Sunless Sea site, here . And any questions or comments about the whole KS process, chip in below!