Thoughts after releasing Divinity:Original Sin and what comes next

Ok,  let’s do this thing. Lots of people have been asking me for numbers and thoughts on the release of Divinity: Original Sin, so here are a few.

Divinity:Original Sin did pretty well. At the time of this writing its Metacriticcritic rating is at 87%, it’s user rating at 89% and it’s been at the top of the Steam charts for most of the summer, occupying the nr. 1 spot for around a month.

It has sold well over half a million units by now– mostly from Steam, with 10% from retail.  ”Break even” has been reached, our debts have been paid and we are now in the profitable zone. While not all of the money is for us as we had private investors on board, the game did sufficiently well for us to envision funding our next endeavors with it, meaning we’re pretty happy about its performance.

Divinity:Original Sin was received well by both players and critics. It has the highest user score of all recent PC games.

So much for turn-based fantasy RPGs not selling, crowdfunding not working and a developer like us not being capable of bringing a game to market without the help of seasoned publishers  .

It’s now been two months since release, so I’ve had the chance to recover somewhat and organize my thoughts.

Releasing D: OS has been a fantastic and rewarding experience, but it’s also been a very tough game to make. There were a surprising amount of hurdles we had to jump and negotiating them wasn’t always easy. I wish I could share my mailbox so you could appreciate some of the finer nuances of some the prose that was directed at yours truly, but alas, confidentiality obligations prevent me from doing so.

The release of D: OS was one big crunch period with all the good and bad that come with it. If the game ultimately did well, it’s because of the outstanding performance of the team when “the going got tough and the tough got going”.

A lot of the crunch was caused by our decision to listen to the feedback we received through our Kickstarter and Steam Early Access communities. While it often was tough to read through all of the criticism, it was clear that integrating the best parts of the feedback would be well worth the effort and improve the game massively. We didn’t hesitate for a minute.

This meant extra delays however, which in turn meant a need for extra budget. Steam Early Access was getting us some money but unfortunately that wasn’t sufficient. We needed to pay back our creditors who were all under the conviction that the game would be out sooner. When, to my surprise, it turned out that they didn’t share our belief that everything was going to be ok and even better if we listened to the feedback, I had to engage in a lot of fun conversations. Between “it’s ready when it’s done” and actually following up on that mantra, there unfortunately lies a big gap that can only be bridged with financial stamina.

I think we would’ve continued development even longer, but when I had to dash to a far away place where lived the one last bank director who still wanted to give us sufficient credit to pay a part of what we owed to another bank, it was clear that we needed to finish. I wasn’t joking when I said it was all in.

When D:OS finally shipped, a lot of us were exhausted. Perhaps surprisingly, but typically, not all of us were happy that it was finished, as there were still a ton of things we wanted to do. Our bodies and our families had different feelings however. They were very happy that it was over and we needed to make quite some amends.

Personally, I was completely worn out, both mentally and physically, and I needed to detox myself from “the game”.  I’d been waking up and crashing down with it for so long that I had a hard time getting used to “normal life” again. My partner took me to Rome the weekend after D:OS shipped and to my shame all I could think of was checking the forums to see if everything was ok. I’m very lucky that she’s still with me.

We worked on D:OS until the very last day before release, and while that in itself isn’t for the meek of heart, it did have some interesting consequences. For one, we didn’t have any review code to share with reviewers prior to release. This meant that it would take several weeks before we’d actually know what the review scores were going to be. It also meant that anybody interested in the game would have to either wait or check what other players were thinking.

I don’t know if there was any correlation between our ultimate review scores and the user reviews, but the latter were really good and when you went to the Steam page on the day of release, it was loaded with over1500 user reviews, 93% being thumbs up. I think that fuelled a lot of the initial success of the game and I also think it made some reviewers pay a bit more attention to the game.

Like (I think) most developers, we were well aware of most of Divinity: Original Sin’s issues before release. However, the freedom the game offers, together with a very small QA team, ensured that it was impossible for us to test all permutations, so we had to take some risks. We quickly patched whatever caused too much grief and while the quick patching wasn’t without risk either (we needed to release quite a few hotfixes after releasing a patch), the vast majority of people could play without encountering game-breaking issues. Our biggest issues were with 32bit operating systems and to this day we are working on improving stability on those.

Our plan is to continue supporting D: OS for quite some time as this is the RPG framework on which we’ll build our next games. We’re fooling around with controller support to see if a big screen version with cooperative play would work well, something I’m silently hoping for as I think it’ll be a lot of fun, more so perhaps than playing coop in LAN with a friend sitting next to you. We’re also improving the engine itself as well as adding a bunch of extra features that not only make D: OS more fun and more friendly to players, but that will also improve whatever our next offering will be. We’re also adding extra content, like for instance the big companion patch, voiced et al, and I imagine that won’t be the last of what we’ll add.

The foreseeable future for Larian (i.e. the next couple of years) is going to see us making further progress in improving our RPG craft and creating dense game worlds with hopefully new and innovative gameplay systems based on old school values.  These last months I’ve been very busy expanding our development force so that we can continue to compete in tomorrow’s market.

As I mentioned in this interview, the current thinking is that we shouldn’t go back to Kickstarter. That’s not because we’re ungrateful of the support we received through our Kickstarter community or because all those rewards caused a bit of extra work, but because I think the crowdfunding pool is limited and it should be fished in by those who really need it. Since we now can, I think we should first invest ourselves and then see if we need extra funds to fuel our ambitions. Only then it makes sense to look at crowd funding. I know several of our backers will be displeased by this, so it could be that we still change our minds, but if that is the case, I do think the the format we’ll use or the way we’ll do it will be different than how we did it for Divinity: Original Sin. (Update: I forgot to mention in the original version of this post that we will be looking at ways of engaging our community sooner in development, but haven’t made any decisions on this)

Let me end this post with a few key observations I made for myself during the development of D:OS.  I hope they’ll help some other developers because they for sure would have helped me had I adhered to them from the very beginning.

In the modern game industry, content is king. It’s a good moment to be a developer, but you have to find a way of doing most of it yourself, without any third party being involved. For each third party you add, your development complexity is going to rise exponentially  because you get extra communication lines and there are no guarantees that throughout development, your agendas will continue to match. We had a lot of third parties involved during development, and it was the root of half our problems.

Content is also king when it comes to the game itself. Bad content means players will have no motivation to invest themselves in your game, and if you find yourself for whatever reason with bad content, cull it. Don’t release it, even if it causes extra delays and it brings you to the brink of bankruptcy. Bad is bad and players will recognise bad. Fix your content first. At some undisclosed point in development, we dumped more than half of what we had. It was the best decision ever, even if did tarnish the end result a bit and caused a lot of extra stress.

If content is the king, polish is the queen. The best content in the world will get low ratings if you have a poor UI and no gloss, or if players don’t understand your systems. These are easy things to say, but they are very hard to put in practice and sometimes you find that you may have to backtrack a lot. Don’t hesitate about this, just do it.

The posts on our forums are veritable goldmines of feedback but beware, for experienced miners only

Listen to your community, but be aggressive in your triage and remember that game development requires enlightened despotism, not democracy. Communities are diverse and as such you’ll get conflicting opinions. You need a strong core vision to guide you through their feedback, and you need to stick to that vision, no matter how vocal they become. But you do need to listen and recognize the underlying causes of problems being reported. Often communities will complain about the symptoms of something that’s wrong and it’s not always easy to discover what the root cause is.

Also remember that the vocal minority does not represent the majority, no matter how hard their claims. The majority doesn’t have time to write thousands of posts. And if you encounter some uncivilized people on a forum,  ignore them. They’re not worth the emotional stress they may cause. You wouldn’t deal with them in real life either.

And finally, iteration pays off. I’ve tried countless things to avoid iteration, but all of them have been failures. At best you can try to reduce the amount of iterations, but you will need to iterate. Make it part of your development ethos; it’ll save you some hardship.

There is of course much more that I could say and I hope I’ll update my blog more frequently, but as it happens, I am pretty busy for the moment. In addition to strengthening our development muscle, we’re starting up our new productions, which is a lot of fun– but also a lot of work. The choices we’re making now will affect us for the next couple of years, so I’m trying to maximise the amount of mental bandwidth that goes into them. I remember all too well some bad choices I made when starting development on D:OS, and I’d very much like to avoid repeating them.

In conclusion, developing D:OS was a very rich experience and I think our entire team matured a lot in the process of making it. We do this job because we enjoy entertaining other people with our imaginations, and when it’s successful, it makes us feel all good inside. We’re very grateful for the many thank you messages we received from our players: it’s these shows of appreciation that fuel us when the night is dark and the task list long, and it’s what makes this job so incredibly cool.

Thanks for reading! There are a lot of topics touched upon in this post so I’m looking forward to your comments.