Of all the minefields currently strewn across indie game development, Early Access has to be the most volatile. Just as likely to ruin as it is to lead to success, there’s been an increasingly frustrated sea change among players over the past few months as multiple projects have imploded or been suspended, developers either leaving projects unfinished or finding themselves unable to meet the promises they’ve made.

The problems come from multiple different avenues, from treading the line between marketing your game on the experience it will be and overselling what it could be, to just launching it far too early, to not properly managing the community you’ve fostered after launching Early Access. These problems are compounded by being relative to each game.

Make sure your game is ready for Early Access

The term ‘minimum viable product’ reads particularly callously when applied to games, but it’s the first and easily the most immediately important problem an Early Access developer faces. The answer will be different for each game, but after talking to Chris Bourassa and Tyler Sigman of Redhook Studios, the creators of the recently released Darkest Dungeon, it's clear there are certainly common considerations

“I think that the core of the game needs to shine through.” Bourassa explains. “If you’re making a game about dungeon crawling and the emotional toll and stress related to that, that has to be in your Early Access build. You can’t ship half of that and then expect people to wait around for the rest of your promise. I think the premise of your game needs to be fully encapsulated and realized. Maybe not fleshed out, but it has to be there.”

Florian Frankenberger, who was brought on to Towns after the original developers burnt out and the community had already soured to the project, has similar things to say. “I think for me a ‘minimum viable product’ is a game that first of all is stable enough to not crash that easily, although there are still bugs. Secondly, it’s a game that can be played and is fun while doing so. All in all I think the player should get a good impression of what the game is and where the journey will eventually go. And all that without getting too frustrated because of all the bugs.”

Another aspect that Bourassa highlights is to consider whether releasing on Early Access is actually useful at a point in development. “Sometimes when you release something early, you get a lot of feedback that you already know that you need to address, so you’re not actually leveraging the Early Access community in a way that makes the most of their playtime. They’re reporting on UI glitches, or wild imbalances, and we may already be aware of that stuff, it’s not new information. It’s more effective to release something that’s a little bit further along, because you get more insightful or interesting feedback. That’s much more valuable to us as developers.”

That’s the ideal situation, where you can delay release to make sure that you’re maximizing the effect of your launch, so that both you as a developer and your players are happy with what you’re putting out. But all that time that you’re delaying you’re not making sales, which is certainly something to be aware of.

Redhook were made very aware of that as they delayed their Early Access release by an extra few months, from October last year to February this, a decision that pushed them to the edges of their financial tether.

“You can really do yourself in by limping into Early Access.” Tyler Sigman explains. “That’s just not something we could really afford to do. So there was a tension between waiting as long as you can, to make sure it’s good, but we still have personal runways, finances to juggle; things like that are tough, so there’s always a pressure to release sooner, but that just wouldn’t have served the interests of the game, so we couldn’t do that.”

Redhook had the advantage of a very successful Kickstarter campaign that gave them a reserve that they could draw on, but that’s not something that’s available to every developer. Garry Newman of Facepunch Studios, creators of the wildly successful Garry’s Mod and the currently in-Early Access Rust, made the latter available from a very early stage.

“Thinking about whether to have open development wasn’t ever something we did, it was just assumed.” Newman tells me. “The decision to go on Steam was just a decision based on the logistics of distributing the game.”

Manage player expectations, and don't over-promise

It’s hard not to see the early availability of Rust causing Newman and his team some problems early on, as it was originally launched as a contemporary of DayZ, complete with zombies. But then, a few months into development, Facepunch removed the zombies entirely. A very vocal portion of the community was outraged.

The problem here was one of mismanaged expectations, not on the part of Facepunch, but just on the part of Early Access as a concept. As Newman puts it: “I think a lot of people think that the main game is done and we’ll be adding free DLC every month, when actually it’s something that will change over time.”

It’s to be expected that a new method of interacting with games is going to bring its own problems, but we’re close to a nadir of regard towards Early Access at the moment; projects like Godus and The Stomping Lands have all but folded in on themselves, collapsing under the weight of promise and the lack of any visible progress towards them.

Steam, for its part, has started to turn the ship, as inexorable as it is glacial. In their trademark hands-off approach, Valve has attempted to nudge dvelopers towards better practice, with the most drastic change being the forbidding of concrete promises as to exactly what and when things will be added to a game.

“They don’t want you saying ‘For sure it’s coming out on this date’, or ‘for sure we’re releasing this feature’,” Tyler Sigman tells me after I ask about how Redhook are handling things with Darkest Dungeon, now that they’ve launched Early Access. “It’s kind of a balancing act because you want to let people know what’s coming without violating that agreement, but the main reason they’re saying that is because they don’t want you selling your game on promises. I think it’s important to talk about what you’d like to do, rather than what you’re 100% going to do. That’s where you get into trouble.”

Towns, for its part, fell catastrophically afoul of this, being one of the first Early Access games, as well as one of the first to stumble in a big way. Selling itself as Diablo-meets-Dwarf Fortress, there was a lot of excitement about the possibilities it was suggesting -- excitement that shifted a lot of copies. When that promise failed to materialize, the community soured.

“In the beginning of Early Access there was this misconception that everything that was listed on the game’s info page was an irrevocable feature of the game.” Frankenberger explains to me. “After all, most Steam customers got this idea from the finished games that were sold on Steam before the Early Access program started.”

Things are turning around for Towns now that Frankenberger has taken over from the original developers, Xavi Canal and Ben Palgi, neither of whom are involved currently. Primarily this is taking the form of increased communication and involvement with the community, something that was a major problem previously.

Communicate with your players, and keep development visible

And it’s perhaps here that things most often go wrong, and where things shift away from traditional development in a big way. As with all things it’s going to be a balance, but it’s not only important to talk to your community, but also to organise development in such a way that what you’re adding to the game is visible to the players.

“I think there’s stuff that’s out of sight and very important, and the stuff that is visible, and we’ve definitely talked about making sure there’s a mix.” Sigman explains. “Communicating with people about what you’re doing and going to do is very important too.”

For their part, the community itself can be an ally to the developer, rather than merely something to appease and keep happy. With Rust’s drastic changes, Garry Newman has often had to weather a vocal minority of angry players.

“In situations like this the community overcomes itself. Everyone starts out ignorant, then a few guys get what the deal is, and take great pleasure in arguing with everyone else. Their number grow as everyone slowly realises which side of the argument is the right side, and eventually it becomes a non-issue because it’s common knowledge. Our job is to give those few guys the ammunition they need to fight the rest of the ignorance.”

As to how to actually do that, Newman has this to say: “Talking to them every week, showing them that we’re listening, explaining in detail any problems we’re having and how they affect them. If you’re a game developer that really loves what you do then talking about it is something that you’re going to take great pleasure in. This is a situation where people really want to hear what you’re doing.”

Frankenberger agrees. “One thing that specifically went somewhat wrong with Towns is communication. Most developers underestimate the importance of proper communication with the community. Every word you say is read, reread and reinterpreted by the community. It’s very easy to quickly post something on the game’s forum and then get people angry at you because they got the wrong impression. It’s important to carefully plan your steps with such an endeavour as developing a game.”

Fundamentally, Early Access is an exercise in managing expectations, both your own and those of the players. There are always things that are going to be nice to have, features that would be ideal if and when you had the time and budget to implement them, but if you create the expectation that they are a given, or even likely, you can potentially dig yourself into a hole that you can never emerge out of. Ambition is laudable, so long as its tempered by an awareness that things don’t always go to plan.

“It’s always going to take longer and cost more than you think. That’s the bottom line.” Sigman tells me when I ask him for advice for developers considering Early Access. “That was true for us too. You need to be somewhat realistic regarding whether you can get there. What Steam is trying to curb is people using Early Access solely as the next meal. But there’s obviously a reality that sometimes, it is. If you think you’re going to barely limp in you have to add contingency on top of that, because it will take longer than you think.”

This sentiment was echoed by pretty much everyone I talked to. The perception of Early Access is that it is a way to start earning from your game far sooner than usual, but the reality is that the model only fits when the game will actually benefit from that public exposure before completion. Harnessing your community feedback, and turning the development into an event for the player, are great ways to cultivate a successful Early Access, and while sometimes it’s the only viable option, the question always needs to be asked about whether it’s the right one.