Why your launch predictions will be wrong. But that’s okay.

When launching a multiplayer game, it’s extremely difficult to predict how many players you’ll actually get. We look at why that’s okay, as long as you prepare.

Before you pop the bubbly, cut the red ribbon, and officially launch your new multiplayer game – there’s one big question you need to answer. How many players are you expecting?

It’s a tough one. The answer determines how many servers you’re going to need and how much you’re going to pay. If you get it wrong, you could have huge queues of players – waiting to join a match. Not having enough servers to handle the load can lead to bad reviews and downtime.

Nobody wants that on their day of celebration. You’ve worked hard for that launch. So you need to predict how many players you’re likely to have. The problem is that those predictions are very, very likely to be wrong. Don’t worry. That’s normal. The trick is to prepare for the best.

How do you predict your player count?

Predicting how many players you’re likely to get feels a lot like staring into a crystal ball and hoping to see the answer. The picture is never quite clear. And it’s easy to misinterpret the signs.

If you’ve released similar games before, you can use those numbers. Previous games can give you a good indication of how popular your game might be. But that only helps if it’s a direct sequel. If it’s a new genre, that data can be misleading.

Instead, preorders are often the best way to get a good benchmark. Or you can create a newsletter and count how many people subscribe. Either method is a good indicator how popular your game might be. But both are only going to show you diehard fans. It’s a baseline number you can extrapolate from, but it’s not going to be your final tally.

However you decide to predict your player count, remember to break your numbers down by region, too. How many preorders are coming from Asia? How many subscriptions are from Europe? How many site hits are you getting each day from North America?

This data will help you get a rough estimate of your likely players, and the percentage difference between the regions.

You’re not going to get it right

Anybody who says they can accurately predict how many players you’re going to get is lying to you. It’s just not possible. You can’t predict a surge of players.

What if a streamer picks up the game? You could easily get thousands more players on a single day, just because of one YouTube video or Twitch stream. This is even more likely if you’re giving out free codes to influencers – which your marketing team is surely going to want to do.

Likewise, if you get good reviews, you could get a surge. Add your game to a Steam sale, you could get a surge. Run a free weekend, you could get a surge. There are just too many variables for you to account for them all. But that’s not to say you shouldn’t attempt to make any predictions at all.

Prepare for success

The predictions you make can help you budget for your hosting and figure out your margins. If you plan for what happens in the most extreme cases, you won’t get blindsided by an unexpected bill.

The trick is to make sure you can be flexible. Look at how much it’s going to cost you if you get half your prediction, if you hit the prediction exactly, if you double the prediction, and what – luck have it – you quadruple your prediction. Can you afford all those cases? And do this for each region, so you’re not surprised if you become a cultural hit in Africa, but fail to capture people’s attention in Europe.

Whatever platform you use for your hosting, you need to make sure it can handle a sudden surge of players. Cloud is often people’s first go-to for this. But be wary. It can get very expensive, very quickly. They often charge for your peak usage. Check your contracts and make sure that you’re able to pay for those peaks, even if they’re shortlived.

Gameye automatically scales, without the cost

Our orchestrator uses multiple server providers from around the world. If and when you get a surge of players, we automatically spin up new sessions immediately. So there’s never an issue. You also only ever pay for the time you actually use, and everything is included in a simple per-minute rate. We’re willing to work with you to figure out a billing system that works for your game. Just get in touch.