Lovecraftesque – a kickstarter retrospective (part 3)

This is part 3 in a multi-part series looking back on our RPG kickstarter, what went well, what we’d change, and so on. Part 1 is herePart 2 is here. This time, we’re talking about how we delivered the product, particularly how we managed the delivery timetable and our costs.

Delivery phase – time

We did ok with keeping to our timetable. We delivered it late, but only by about 1.5 months. In the end this came down to a single delayed workstream which we couldn’t have known in advance would hit problems, and which was too critical to the project to work around. Seen in that context, 1.5 months delay is not too bad – though we would aim to get that down to zero next time.

We did several things to keep the project on track, and they largely worked. First, we calculated that there might be delays, and included that in our project timetable. We’d seen so many Kickstarters delivering many months or even years late that we were trying very hard to be realistic or even pessimistic about our timetable. We actually doubled the time estimate we’d been given by one of our contributors, because we felt it was too optimistic.

Second, we ensured that everything that could be done early was. So we wrote the game in advance of kickstarter launch. We had planned all of the art and layout in advance, and initiated work on them as soon as it was apparent we were going to fund. We had the layout completed for everything we could, even while waiting for some material to come through. In short, we ensured that we could concentrate on delivering the difficult stuff, by getting the easier stuff out of the way fast.

Now before I mention the next factor, this is a good point to recognise that people – and you should include you, the designer, in this category – are a major risk factor for any project. Everyone has a real life that can distract and delay you or even take you out of commission. Health issues can spring up, family tragedies… these are realities in an industry where almost nobody does this as their day job, so we’re all trying to squeeze out work in our spare time. (In fact, our second child was born right in the middle of fulfilment – though we did at least get 9 months notice of that!) I hope it’s clear that I don’t regard this as something to blame or shame people over: it’s something that you as the project manager have to do your best to mitigate. It turned out to be a significant factor for Lovecraftesque.

So, third, we had said up front that we weren’t guaranteeing any of our stretch goals. We would drop a stretch goal or deliver it electronically rather than let its non-delivery delay the project as a whole. I think that was a good thing to do. We didn’t have to drop any, but because we’d said we might, we didn’t need to feel too bad about having to take on authorship of one stretch goal ourselves, supported by a couple of additional consultants that we took on at short notice. We didn’t get any complaints about this, and we think the resulting product was high quality and – crucially – didn’t delay publication.

Fourth, we were very active in managing our contributors. We set deadlines, we reminded people about them, we nagged them if they were late, we negotiated additional time for those who needed it. We worked with contributors to make sure their work meshed with the vision for Lovecraftesque, giving comments and drafting assistance. I hope and believe that this was done in a supportive way, to get the very best finished product.

There is a final thing to recognise in our delivery timetable. While the books were ready and delivered to many backers in October, some of them went missing for about 7 weeks. We don’t know exactly what happened. The records suggest that books destined for the US were shipped from Britain to continental Europe, and then – for some reason – send to Budapest, where they sat for a long time. They were eventually (most of them) sent on to their US recipients. But this caused us a lot of stress and worry, since for a long while we thought they’d simply vanished. And it meant that some of our customers received the product 3 months late rather than 1.5 months. My guess is that this could happen to any project (it certainly seems to be a common problem, watching other projects). In future we would consider paying out on more expensive shipping to allow us to have greater confidence about this.

We’d originally chosen Lightning Source as our printer because they had branches in the UK and US. We planned to ship our US backers from the US branch, which would probably not have been subject to delay in this way. It was a great plan – but one which we had to abandon because of cost. Fluctuating currency values, which we’ll return to later, made the US print branch unviable for us. Post-kickstarter, we’ve offered our US customers the option to pay a (small) premium to get the product printed in the US and therefore delivered faster. Perhaps we could offer that as an option in future kickstarters – though as against that, this strikes me as potentially confusing for customers.

Image of a sleeping person in bed. Behind their back we see luminous creepy-crawlies coming out of their phone and climbing into their ear.
One of the illustrations from Lovecraftesque

Delivery phase – cost

Next, let’s talk about cost. Our carefully costed project was almost exactly on-budget. Our costings – which all included error margins for inflation, currency conversion and suchlike – came in literally on the money, pretty much every time. And yes, that includes shipping: as mentioned earlier, we shipped directly from the printer at very reasonable cost. Runaway shipping costs were my biggest fear throughout the project, and we dodged that bullet thanks to a forensic examination of the costs in advance.

And yet, as I mentioned earlier, we spent our entire 10% contingency fund. Why? Well, the single biggest factor was: Brexit. The value of the pound dropped by a total of 20% between the launch of our project and delivery. 10% of that was pre-referendum, but in retrospect was probably reflecting uncertainty about the referendum result. The rest came after. Most of our stretch goal writers, plus both our artists, were paid in dollars. The resulting exchange costs came to about £600.

The other major factor was to do with the Special Cards. Basically, we underpriced them. Or to look at it another way, we underestimated demand for them. Let me unpack that. From examining previous RPG kickstarters, we thought about half of people who purchased the physical book would want cards as well. In reality, well over 80% did – and in response to backer demand we created a “PDF + cards” tier that we hadn’t planned on. Now, the cards had a very low profit margin on them, which we’d taken into account in our planning – but when we sold a lot more of them than we anticipated, that cost us a bit. Plus they were priced in dollars, so this came together with the Brexit factor in a bad way. After the kickstarter we raised the price of the cards from £5 to £8, because of this.

The cards also added complications to delivery. They were printed and shipped separately to the books (Drivethru Cards is a separate printer from DrivethruRPG). They were an extra bit of admin, an extra delivery risk, one more thing to track and worry about. So, with the above… I’d dearly like to avoid using cards in future. Or rather, I’d like to try and stick to one physical product: cards or a book, probably not both.

So with all this, plus a handful of much smaller things, our contingency fund was spent in full – plus £5 over, to be precise. Sigh. At least we had one. But despite this, we did not make a loss. We included payment for ourselves in our funding goal, and increased it with each stretch goal – something I would always urge you to do, if you’re planning to kickstart something – and so we actually made a very respectable amount of money from the kickstarter itself. And of course, we never had any money-related problems delivering the product.

By the way, I should also mention that we used Backerkit to do our customer survey. This allowed non-Kickstarter backers to purchase books after the campaign ended and enabled existing backers to upgrade their initial pledges. True to what Backerkit estimate, we sold a sizeable additional chunk of books through them. A good choice, which I’d recommend to others.

Lovecraftesque – a kickstarter retrospective (pt 2)

This is part 2 in a multi-part series looking back on our RPG kickstarter, what went well, what we’d change, and so on. Part 1 is here. Part 3 is here. This time, we’re talking about how we worked to support diversity and inclusion, and tackle Lovecraft’s bigotry; and how we ran the Kickstarter campaign itself.

Diversity and inclusion

One of our goals for Lovecraftesque was to be an exemplar on diversity and inclusion, in as many ways as we could. Aside from this aligning with our politics generally, we specifically wanted to punch Lovecraft’s bigotry in the face – to make the kind of game he would have thoroughly disapproved of.

We did this in a number of different ways:

  • We directly addressed Lovecraft’s racism and the attitudes towards mental illness that are embodied in his writing and (to a much greater extent) Lovecraftian roleplaying games. We had written some basic guidance on this ourselves, but included as stretch goals more detailed work on this, which (once funded) we included in the game book.
  • We included in the art specification a requirement that the art be diverse, showcasing characters who are female, ethnically diverse, LGBTQ and disabled. We specified ratios for these characters to deliberately put white dudes in a minority in the book’s art. We also asked our artist to avoid casting these characters as weak, submissive or sex objects – although the nature of the game meant none of the characters were kicking ass, as all were potentially going to be victims of something nasty.
  • When advertising for artists we specifically asked PoC and LGBTQ folks to put themselves forward. When approaching individual artists and writers, we aimed to draw on a diverse pool, again keeping white dudes in a minority. We weren’t sacrificing quality to do this – there is a great pool of diverse talent out there.
  • We paid all of our contributors the same highly competitive rates – well above what we understand to be the industry average – because we’d like roleplaying design and writing to be something people get properly rewarded for, especially for people who don’t have the privilege of a rockstar reputation (i.e. most of us).
  • Although we didn’t exactly set out to do this, the diverse team we’d recruited was probably directly responsible for the very diverse set of scenarios that came with the game – from those with a fairly traditional Lovecraft-ey setup to stuff HPL would never have dreamed of.

To judge by the enraged response of some less reconstructed gamers, and those who affect to be tired of political correctness, we succeeded in branding the product as diverse, inclusive and tackling bigotry. I believe we did more than that – we embedded those values in the game itself.

One of the illustrations for Lovecraftesque: a priest stands amongst the pews of a church, comforting a visibly stressed man, in whose lap is a notebook filled with drawings of a scary figure. In the background, the same scary figure is just visible pressed up against a stained glass window.
Did I mention I love our art?

The Kickstarter

Thanks to Kickstarter preserving everything in aspic, you can go and look at our campaign page now if you like. We tried to hit all the information someone might want to understand the product, get enthused and feel confident in us. So: a clear pitch that highlights what the game feels like to play, what makes it different, and what makes it fun; a clear explanation of what you’ll get if you back; a simplified version of the game for free download, to try-before-you-buy; samples of art and layout work; a detailed explanation of how we planned to spend the money; and an explanation of the risks including how we planned to mitigate them. Plus a video, because apparently that’s a good thing to have (I personally am not at all motivated by videos, but statistically I’m in a minority apparently). We got feedback from friends and more experienced people on the page prior to the launch and made changes in response. You know, this stuff isn’t really rocket science, but you do need to check these boxes to maximise chance of success, and we did.

We worked hard to publicise the game from a relatively weak starting position. We had established a presence in Indie G+ circles, over a period of 2-3 years, so we weren’t exactly invisible. Even so, as far as I can tell, we weren’t well known, nor did we have (much of) a following. So we wanted to counteract the newcomer disadvantage.

One way that we did that was by approaching people who were better known than us for stretch goals. This conflicted a bit with another goal of ours, which was to have a very diverse stretch goal team. Early on we deliberately went for an approximately 50/50 split between people with star power and other designers – all of whom, to be clear, we had chosen primarily for their talent and the fact we admired them. We also leaned towards people who seemed interested in horror gaming. I think the presence of both of these groups will probably have helped to generate interest from a broader community than we could have reached ourselves. The “star” writers will have reassured people who like reliable industry quantities. In addition to helping fulfill our mission to create a diverse and inclusive Lovecraft game (see above), the broader set of writers probably attracted backers who like to see a diverse team and who were fans of the somewhat more niche indie game design community.

Another thing we did was to approach a wide range of podcasts and blogs. In retrospect, I wonder if we actually undersold ourselves here. Having seen a friend’s recent media engagement plan, it was significantly more wide-ranging than ours. Possibly we didn’t reach as broad a roleplaying audience as we might have. But that’s not to underplay the level of activity – and quite frankly it’s hard to imagine how we could have done much more than we did, given the stage of life we were at (tiny children limited what we could realistically do). We are enormously grateful to people who hosted us on their shows or interviewed us for their blog, and I strongly suspect that this had an impact on our support – though I can’t prove it. Also, talking to people about our game was fun!

On top of that, we managed to attract interest from some big-name websites including the Mary Sue, Boing Boing and (after the campaign) Geek & Sundry. It’s still a bit of a mystery how we did this, other than having the bare-faced cheek to ask. I suspect the concept of a Lovecraft game that tackles diversity and inclusivity head-on helped. With that said, I saw precious little evidence in our referral data that these websites had an impact on our sales. We had google analytics set up on our kickstarter page, showing where people viewing the page had come from, together with kickstarter’s own analytics saying where our actual backers came from. So for example, Boing Boing got us a whopping 600 website referrals over the course of the campaign – but to judge from Kickstarter’s analytics, zero backers. On the other hand, the Mary Sue got us 160 views and 17 backers. These media splashes widened our reach beyond our wildest dreams, and probably did get us some backers (after all, some may have come back to the page later on, rather than being directly referred by an article). But it didn’t generate the deluge you might imagine, perhaps because they mainly extended our reach to people who weren’t that interested in roleplaying games – and so less likely to actually buy a copy.

While we’re on the subject of where backers came from… where did they come from, anyway? About 250 of our backers had got there from Kickstarter itself: the search page, the “discovery” pages, other campaigns, and so on. That’s nearly half our backers. We also got a lot from G+ and Facebook; some of these will have been friends, but not all. Again, not knowing if people were coming back having seen the campaign through a link but then looked it up later on, means we can’t be confident how they originally found out about the project.

We aimed for a steady flow of activity during the campaign, including stretch goal announcements, interviews and so on. We tried to strike a balance between regular updates, keeping things ticking over, and not overloading our backers (and social media feeds) with constant Lovecraftesque stuff. Again, this didn’t really have a visible impact on our stats, which conformed closely to the standard pattern, i.e. a big rush at the start, a lot of people jumping on at the end, and a pretty steady (low) rate of uptake in between. Again, we’re enormously grateful to everyone to shared our campaign and contributed to the buzz by talking about it, and I assume it had some impact. But nothing we did showed up in our stats particularly.

So… did any of this publicity work help? I mean, in a way it must have. Clearly, if you don’t talk about your game, nobody will know about it, so nobody will buy it. But despite collecting lots of data, there is very little evidence on which of this activity had an impact, and which didn’t. So I can’t say we really learned much, other than that the campaign obviously succeeded, so we got something right.

One thing we did not do is advertise. I’ve noticed recently some Kickstarters showing up as adverts in my Facebook feed. And I know at least one designer who has had some pretty good sales figures (not on Kickstarter) from advertising their RPG that way. We did consider doing this, but decided that spending a non-existent budget on advertising probably wasn’t the best plan. But I do wonder, having seen other people’s success, whether this was a mistake. It’s certainly something we’ll consider for the future.

One final interesting thing to ponder is the grabbiness of the game. People say that you can sell roleplayers anything with Cthulhu on it. We were in the weird position of effectively saying “Come and buy our Cthulhu game! Surprise, it doesn’t have Cthulhu in it, and it doesn’t work like a normal Cthulhu RPG, and we spend a big chunk of the text criticising and addressing HPL’s bigotry.” We know that at least a few people bought the game having not looked beyond the Lovecraftian brand (though we were excruciatingly clear on the campaign page what we were selling) and were then disappointed when they got the product. We could equally conjecture that some people were put off by the Lovecraftian brand, not having registered the ways that the game addressed the things that give Lovecraft a bad name. Still other people may have been drawn in by Cthulhu and not backed us when they saw what the game was really about.

In the end, we just don’t know how these things interacted. We had a very successful first Kickstarter, and we’re delighted by that.

In part 3, we’ll look at how we delivered our successfully funded kickstarter.

Lovecraftesque – a kickstarter retrospective

Just over 2 years ago, we launched the Lovecraftesque kickstarter. If you didn’t follow it at the time, it was a success – we raised £15000, over triple our funding goal. As we’re slowly moving towards a position where we might kickstart some more games[*], it seems timely to talk a bit about what went well, and what went less well.

It turns out there’s quite a lot to say, so this is part one in a multi-part series. Part two is here. Part 3 is here. This part looks at the planning we did ahead of the kickstarter.

Design phase

We started writing the game a good year before the kickstarter. We had a clear concept, and clear design goals.

We were very fortunate to get a lot of interest from potential playtesters, many of whom actually playtested the game. We asked clear, focused playtest questions. As a result, we got great feedback which allowed us to refine and streamline the game. I think this is visible in the quality of the design.

We already had a completed draft ready when we launched the kickstarter. I suspect that knowing the game was written was a selling point, for a debut project. It also meant that we had one less thing to worry about in the delivery phase.

Planning phase – printing and shipping costs

We put a great deal of thought into planning the kickstarter.

We identified potential printers, and took quotes from a few.  We plumped for delivery through Drivethru‘s printer, Lightning Source aka Ingram Spark. (We went direct to the printer rather than through Drivethru because the prices were slightly lower.) We’ve been delighted by the quality of the books they produced, and the price was reasonable – albeit a lot higher than if we had done a print run. They ship directly to customers, which is keeps things simple for us logistically.

In retrospect, however, we may have got this wrong. We now know that we could have got a significantly lower price per unit than we actually got if we had gone with a print run instead of POD. The sample of non-POD printers we looked at didn’t appear competitive, but perhaps our sample was too small, or maybe we should have asked for recommendations. I don’t want anyone to get the wrong idea; for smaller numbers, which is what we had anticipated, Print On Demand makes a lot of sense. It reduced our exposure to the risks entailed by up-front expenditure. It allowed us to produce as many copies as we wanted after the Kickstarter, because we didn’t have to decide on a number of copies in advance. However, knowing what we know now, we probably could have saved £1000s with a print run. This would be particularly true if we had gone for a single format rather than offering both softcover and hardcover.

Specifically: we priced our initial funding goal for just over 100 physical copies. In practice, we funded at triple our goal, including 400 physical copies. But perhaps more importantly, we underestimated the number of copies we might sell after the campaign. As of now, we’ve sold 1100 copies, and are still selling a steady stream. That’s easily enough for a print run. This would have entailed an up-front cost. But the cost reduction per unit (excluding shipping) would have been huge, at around a £5 saving per unit. A print run of 1500 would not have been much more expensive than producing 400 copies by POD. This excludes shipping, so may overestimate the benefits of a print run: but the lesson is that you should think ahead and at least consider whether a print run might work out better for you.

On which subject: the huge plus of using Lightning Source was that their shipping costs were very, very low. We had done our research and got the message: shipping can be expensive, especially international shipping. Lightning Source delivered to anywhere in the world (with a very small number of exceptions) for about £2-3. If we had a comparably successful product in future, this would be the one factor that might decide in favour of POD over a print run.

Planning phase – art and layout costs

For both art and layout, we researched and selected a shortlist of 3-4 individuals to approach for a quote, based on work they’d done before. We actively solicited a diverse pool of artists to shortlist from – something I’ll cover in more detail later on. For layout artists, this was based on looking at games we owned; for illustrators, we looked at the artist’s online portfolio. We selected partly based on price, and partly based on how much we liked the artist’s work. They were all great, or we wouldn’t have looked at them. We are really happy with the work our artists did.

One of the illustrations for Lovecraftesque: three children and their dog swim in a deep blue waterhole. A massive alien eye lurks at the very bottom of the picture.
I love commissioning art SO MUCH.

We set specific goals for book length, and number of illustrations, for our basic funding goal and each of our stretch goals. took into account the extra printing costs that this would mean, and got specific quotes for each goal, including for extra illustrations, scenarios and so on, from our artists, so that we knew in advance exactly how much we would be paying, even for stretch goals that we hadn’t set at the start of the project.

Planning phase – other stuff

We had looked at a dozen other (completed) RPG kickstarters of a range of types, to get an idea of what the likely distribution of rewards chosen would be. We had a good idea of what percentage would be PDF-only, softcover, hardcover and so on: and these predictions were largely right. This helped us to ensure we were setting the correct funding goal, because the amount of money you keep varies by reward level (simply put, PDF has no print or shipping costs, so you keep 100% of the money, while physical copies vary in their costs).

We sought advice and views from lots of people in the run-up to the campaign. The biggest source of advice was Google Plus, where people generously gave their views on a wide range of issues we were worrying about – including many experienced people sharing their knowledge of how to run a kickstarter. We were super-grateful for all the advice we got, which hugely enhanced the research we ourselves had done.

We approached a handful of designers we knew and admired to write scenarios and essays for our stretch goals. We were gratified by the response – almost all said yes. We had set ourselves a goal to offer rates that would be significantly above the market rate for RPG writing, which may have been a factor in people accepting work from a couple of randos they’d never heard of. We set a word limit for our stretch goal materials, and stuck very closely to it, which meant that again, we knew in advance how much we would be paying for this work.

Finally, we had a contingency fund of 10% of our total funds raised, to cover stuff that might go wrong. This was a very good decision that we took early on, and – by fluke – we spent it almost exactly. (I’ll mention here that we’d planned to give an equal share of any remaining contingency money to ourselves and our artists. It didn’t happen, because there was nothing left, but we still think it was a pretty good idea, and will likely take the same approach in future campaigns.)

Next: How we ran the campaign itself – and how we supported diversity and inclusion.

[*] Yes, plural.