While I try to write candidly, I don’t normally talk about every motivation that we have when making decisions. I feel that this is a case where it’s important that we lift the curtain a bit and explain all the factors that influenced this.
Action RPGs are all about finding items, levelling up characters and accomplishing difficult challenges, all while comparing your progress to your friends and feeling that you found better stuff earlier than they did. The most exciting time for this is right after the launch of a new league, because it comes with a fresh economy to play in. Players arrange time off work, queue to log in right as the league opens, and then push really hard to get established in the league before other people can. The first days or weeks of a new league are by far the most exciting time for players, and this is reflected in our metrics that track player engagement. There is a massive spike at the start of a league, which gradually drifts down over the 13 weeks before it spikes up again for the next one. We have designed our entire business model around this, by doing our content releases at league launch, our marketing at league launch, releasing our mystery boxes, supporter packs and new microtransactions at league launch, etc. It’s incredibly critical that league launches go as well as possible. This is also why we have our launches on Friday afternoon/evening (depending on whether you’re in America or Europe) – to maximise weekend time available for players’ first push.
So what impact do other games have on this? So far we have historically been really lucky to almost always avoid directly clashing with the release of other major games. However, in late 2019, we released the Blight league less than two weeks after WoW Classic’s launch. This had a massively negative impact on player numbers, play hours, Twitch views and revenue. It was a bit of a disaster, and quite a big learning experience for us. The truth is, for a lot of players, if they miss the initial launch of a league, they’ll probably skip it or not take it anywhere near as seriously as if they managed to make good progress and get established in the early days or weeks. It’s vitally important to have a clean release. As a side note, this is why we also have a lot of implicit motivation to make sure our releases are as stable and bug-free as possible. Bad launches have very large ripple effects that could threaten our ability to sustain the game.
Cyberpunk 2077 is a massive game. A lot of people are going to play it, and they’re probably going to play it pretty hard. By looking at the stats from services like Steam, you can see the player engagement curves for large releases like this, and it’s easy to see that there will likely be several weeks of absolute dominance while everyone enjoys it. I understand that you may feel that you can run through a single player game in a weekend, but realistically gamers as a whole are going to be very occupied for at least a few weeks.
Prior to yesterday’s announcement, we were somewhat comfortable with the relative timing between Cyberpunk’s previous release date and our target date for 3.13. There was a three-week gap, which was closer than we’d like but probably enough time to not existentially threaten 3.13’s release.
They then announced a delay of their release to within 24 hours of our launch. I understand that if you don’t plan to play Cyberpunk, you may feel that it’s an overreaction for us to want to move out of its way. I am deeply sorry about the inconvenience and disappointment. But the commercial reality is that if we kept our original release date, we would be fucked. We believe we’d lose at least half our players, almost all of our streamers, a chunk of our developers, and we honestly wouldn’t blame them. Launching at the same time as a game of that size is not feasible.
Okay, so can we launch earlier in December? Well, not really. As we mentioned last week, 3.13 has been scoped carefully to avoid development crunch and quality issues at launch. Trying to finish it faster would reintroduce both problems. Also, how many weeks early would we have to launch in order to actually avoid getting trampled by Cyberpunk’s release? So releasing early is by far the hardest option.
Alright, but what about later in December? So this issue here is Christmas. In New Zealand, it’s summer in December, and Christmas is typically when people take their annual leave to spend time with their families and friends in the sun. It’s very common for some companies to work 11 months of the year and have all their time off in late December/early January. While we encourage our staff to take breaks throughout the year, there’s certainly a strong bias towards people choosing to be away in late December, and that’s not something that I am going to ask them to give up.
Launching one week later than we planned, on December 18, means that almost all of our staff would be away immediately after launch, preventing us from fixing post-release problems, doing the console launches, supporting Garena/Tencent for their launches, etc. In addition, one week doesn’t really get out of the time window where a lot of people will be playing Cyberpunk. We also can’t launch on December 25 or January 1 for obvious reasons.
So that leaves January 8 or January 15 as valid launch dates (as explained before, hitting a weekend is critical for the launch to go well). A lot more team members are back by the 15th, so that is the date we are likely to announce. I want to say to my team: “You guys had a difficult year. Go have a great vacation, come back recharged, and we’ll launch a product we’re confident in”. To me, this is possibly as important a reason as trying to avoid having our numbers crushed by another game. It’s a very hard call to make, to delay a release by this much, as it’s a huge revenue/schedule hit. In this case the delay being essentially forced upon us does mean that we get some other benefits.
We usually waste a lot of time each cycle preparing marketing materials from a less-than-half-finished build of the expansion, as they have to constantly be updated as the expansion changes. With this schedule, we can make the expansion good first, then prepare its marketing/announcement. It also gives us a month more time for quality assurance, and probably means that we can actually use our Alpha realm for its intended purpose. Alpha testers aren’t allowed to talk about what they see on Alpha, but the state of the game at that point is usually a bit of a shitshow. I think we only managed to get one actual league up and playable this year before its release. That will change if we have this extra month between completion of the expansion and its release.
For the other 50% of Path of Exile players who read all of the above and are still sad that 3.13 isn’t coming out in December, we are trying very hard to prepare interesting events for everyone to play. I’d much rather try to underpromise and overdeliver here, but we are kicking around some ideas. Bex is campaigning very hard for us to work out how to do Endless Delve, for example.
I’m really sorry that this decision has disappointed loyal Path of Exile fans who are excited to see what we’ve been working on for our next end-game expansion and were looking forward to spending their holidays playing it. While I feel that we picked the best of several bad outcomes. While I have been mentally feeling a lot better after committing to this decision yesterday, I still can’t get over the thought that we have let you guys down. I promise that we will work hard to make sure there’s plenty of fun PoE stuff going on over the holidays, followed by a kickass expansion.
