The cadence trap

There is a moment, somewhere between four and eight months after a live game launches, when a room full of intelligent, well-compensated adults collectively discovers that content is finite. This arrives as a genuine surprise to them, in roughly the way that gravity arrives as a genuine surprise to someone who has walked off a cliff while looking at a clipboard.

The team did not run out of ideas. Teams essentially never run out of ideas. Ideas are the cheapest substance in the known universe, available in industrial quantities, usually at two in the morning, and frequently involving a boss fight on a moving train. What the team ran out of was structure.

The pile

Here is how it happens. Before launch, you build a stockpile: a large quantity of content sitting in your build, radiating the comforting illusion of permanence. It looks enormous. You can stand next to it with a coffee and think, quite reasonably, that nobody could possibly get through all of this.

They get through all of it.

They get through it with the unnerving efficiency of a swarm of locusts that has read the patch notes, compared drop rates in a shared spreadsheet, and optimised a route. What you budgeted for eleven months, a meaningful number of players will finish in five weeks. Then somebody will publish a video explaining that your game has no content, and the video will be forty minutes long, which is longer than most of your quests.

Every live game I have worked on has felt the pull of this. The ones that held up were the ones that noticed before the servers went live, rather than during the incident review, at which point noticing is less of a strategy and more of a eulogy.

Design the cadence, not just the content

The unit that matters is not the quest, and it is not the mission. It is the season: not a pile of things, but a shape that repeats, with room inside it for the things to change.

This is the sort of distinction a consultant makes in order to justify an invoice. Inconveniently, it is also true. When the shape is stable, the team gets a rhythm to build against and the player gets a rhythm to return to. When only the contents are planned, every season arrives as a fresh emergency, and emergencies are expensive, badly designed, and extremely bad for the humans assembling them at 11pm.

A season that survives contact with an actual audience tends to have three things:

• A spine that does not move. Players should be able to tell what a season is and when it turns over without consulting a wiki, a Discord, or an oracle.
• A layer that does move. There must be a reason to come back that was not there last time. "We have added the previous thing in a new colour" is technically a reason. It is not a good one.
• A floor that is always there. Someone will return three weeks off-cadence having missed everything, and they need somewhere to land that is not a locked door and an apologetic sign.

The tell

You can usually hear this coming in the language a team uses, which is convenient, because language is free and playtests are not.

When people start talking about filling the next season, the pile has already won. Filling is a thing you do to holes. When people talk about what the season does, for the player, for the economy, for the poor soul returning after a long absence, then the system is doing the work and you can relax by approximately four percent.

None of this is a substitute for good content. You still have to make the thing good, which remains stubbornly difficult and cannot be solved with a spreadsheet, although many have tried, and their spreadsheets are still out there, haunting shared drives, quietly summing columns nobody reads. The craft of the individual quest matters enormously.

The cadence just decides whether anyone is still around to play it.