I can't tell if my browser strategy game is finished. I don't have enough players to know.
TL;DR: I’ve been shipping sizable features into my browser strategy game every two or three days, partly out of a fear that if I stop, the people playing will leave. The game isn’t stable yet and I know it. I’ve got current features to finish off and properly test before I’d call any of it a version 1. But talking it through with a friend, I realised I can’t actually tell how good the rest of it is, because you can’t read a multiplayer game with a handful of daily players on it. The people on it now aren’t testers I can ignore once they’ve served their purpose. They’re early players helping shape the game. The harder question is what the game wants to be, and I need more players on it before I can answer that.
I can’t tell if the game is good enough
Here’s the question I keep circling. The dynamics of a multiplayer strategy game change completely once there are 100 people playing it every day. With a handful, some things you can read fine and some you’re just guessing at.
The things I can read are the visual ones. The world is very visual, so I can watch players find each other, settle near each other, raid each other. That part I can see happening. What I can’t see is whether the whole thing holds up with real pressure on it, whether the systems behave when a hundred people are all pushing on them at once, and whether it’s fun enough to keep someone who isn’t me. That only shows up at scale, and I’m nowhere near scale.
I’m not flying blind on retention, though. I’ve got the numbers, and they’re sobering in the way early numbers always are. Here’s where players who’ve left actually dropped off, by how many days they survived from signup to last activity, over the last 90 days.
So the median dropped player lasts 6.7 days. The biggest single group goes in the first week, which is the first session and the day or two after it, and that’s where most games lose people. But notice the long tail. There’s a real cluster surviving past two weeks, past a month, and a few past two months. People who stick around that long are getting something out of it. The shape tells me the early game is leaky and the back half works for the ones who reach it. What it doesn’t tell me is whether I could close that gap with onboarding, or whether the game just isn’t for most of the people landing on it. I can’t separate those two with the numbers I have.
And none of it improves on its own. To get to 100 daily players I need more traffic into the top of the funnel, better conversion from visitor to signup, a first session clean enough that people stay, and fewer of them gone by week two. Every one of those has to work, and right now I can’t prove any of them do, because there aren’t enough people on the game to read the signal cleanly.
The idle-player change I did on purpose
The clearest example of how one decision spreads is what I did to idle players, and I want to be honest that I did it deliberately.
We had a spike of about 400 players join from one source. A lot of them signed up and left within minutes. So the map filled up with dead accounts almost overnight. I thought it’d be fun, and useful, to let active players take those idle islands quickly and clear the dead weight out. It worked for that. The problem is the rule couldn’t tell the difference between a player who’d quit on day one and a player who’d just taken a week off. So someone who went away for a week came back to find their islands gone. I’d get an angry message, and they usually didn’t come back. Others would log in, see they had nothing left, and log straight back out without saying anything at all.
A friend put the fix simply. At this early stage you need more bots attacking, or you punish the casual players too hard. The pressure has to come from something that doesn’t care who logs in, not from the few real players who happen to be awake and looking for easy ground. So I’ve been pulling the Black Tide, the NPC faction that takes over abandoned islands, into that gap. The trick was making it bother the right people. It now goes after islands with lumen. Lumen is a resource your church produces and stores, a slow trickle that funds the religious side of the game, so a player who’s built their church up tall has invested real time in something slow. That player is almost always an active one, and they’re the ones I want the Black Tide pushing against. Someone sitting on two or three islands and playing casually won’t have much lumen, so the Tide mostly leaves them alone.
That’s the whole thing in one feature. One decision about colonisation cascaded straight into a bigger one about who the game punishes, and then into a third about what the Black Tide is even for. I didn’t expect making a good game to be this hard. Multiplayer cooperative play especially. Every knob you turn moves three others you weren’t looking at.
Shipping out of fear
There’s a worse habit underneath all of this, and I’ve only just named it.
When you have people playing already, and you watch some of them drop off, you start to feel like you owe them. Like if you don’t keep adding things, keep making it better, the ones who are left will leave too. So I’ve been releasing sizable features every two or three days. Not because a plan said to. Because of a quiet fear that standing still means losing people. That’s the startup brain talking, the same constant worrying that I thought I’d left behind years ago. It turns out it comes back the moment real people depend on the thing you made.
The fix isn’t to stop building. I’ll keep going, finishing the current features, tweaking, balancing, adding more events into the events system that’s already there. The fix is to stop building out of fear, and to work out where the game is actually going before I bolt the next big thing onto it. Adding faster was never the answer. Knowing what I’m adding toward is.
The players I have are shaping the game
The reframe that helped most was getting honest about who’s on the game right now.
These aren’t testers I run through a checklist and forget. They’re early players, and they know they’re early, because I message a lot of them myself. They tell me what’s annoying, they show me which features nobody touches, they argue with me about balance. They help me a hell of a lot, and they’re exactly the right people for this stage of the game. The mistake I was making was treating them as the audience I’m building the finished game for. They’re not that. They’re the people helping me find out what the finished game should be.
The friend I was talking to laid out the fork. You either keep chasing the players you have, or you focus on the next, bigger segment. I think the next real version, the one I’d call an actual release, is for a more casual player than the ones here now. Different person, different first session, different idea of what playing even means. You can’t build for both at once, and I’ve been quietly trying to.
The game doesn’t know what it is yet
The honest part underneath all of it is that the game hasn’t decided what it is.
I took a spreadsheet game, the old browser kind where you stare at numbers and timers, and turned it into something with a world and a story and a map you actually look at. That’s genuinely a cool thing to have done. But it isn’t leaning hard in any one direction. It’s not fully the slow, social, check-it-twice-a-day game, and it’s not fully the chaotic war game either. It’s somewhere in between, and that in-between is hard to market, because you can’t say in one line who it’s for.
I can see now why games take years to build. It’s fun right now, but fun isn’t enough. It needs a payoff, a reason the long tail keeps coming back beyond the novelty, and that’s the part I haven’t cracked. I’ve got ideas for it. What I don’t have is enough players to tell me which ones are right. I need more marketing, more people playing and attacking and poking at every corner, because that’s the only way the weak spots show themselves.
The Black Tide is quietly doing some of that work for me. It attacks constantly and across the whole map, so it walks straight into the bugs, the broken edge cases, the wording that doesn’t read right, the events that don’t fire when they should. It’s the most active player on the server, and it doesn’t get bored or polite about what it finds. But it can only test the machinery. It can’t tell me whether the game is any good, and for that I still need people.
So I keep at it. Not because I play games, I don’t really. Because it keeps my brain working, it keeps me challenged, and it pushes what I can do with AI right out to the edge of what’s possible today. The game being unfinished isn’t something to feel bad about. It’s just where it is, and for once I’m fine sitting with that instead of patching over it.
Inselnova is a free browser strategy game you can play in any browser. Have a look.
