Most of us here ship features rather than stories, but branching narrative is a state explosion problem, and that part is very familiar. Here is what actually holds up when a small team tries to give players real choices.
Why Full Branching Is A Trap
If every choice creates two paths, five choices give you thirty two distinct routes, each needing its own writing, scripting, testing and, if you have voice, recording. The cost curve is exponential and the QA surface is worse, because a bug on route 27 is one almost nobody will find before your players do.
The result is predictable. Teams either ship branches of wildly uneven quality, or they quietly collapse them back together and hope nobody notices.
What Players Actually Notice
The useful finding is that players almost never replay a game to verify that their choices mattered. They judge in the moment, on three signals: did the game acknowledge the choice immediately, do characters treat them differently afterwards, and did the choice pay off in a way that felt earned.
A system that delivers all three feels deeply responsive even when the critical path barely moves. A system that skips any of them feels mechanical even when the branching underneath is genuinely complex.
Narrow Branching As An Architecture
The shape that survives contact with a real schedule is a mostly linear spine with a small number of variation points. Two or three real decisions, each with two options, four to eight distinct endings.
Implementation wise this is a flag store plus a reaction layer, not a graph of scenes. Choices write to a small set of persistent flags. Dialogue, ambient lines, item descriptions and NPC reactions read those flags. You author one spine and a set of conditional overrides, so content cost grows additively rather than exponentially, and new reactions can be added late without touching the plot. Ink, Twine and Yarn Spinner all model this well.
The Web Specific Constraint
Browser sessions run roughly two to fifteen minutes, which kills the slow burn structure outright. The hook has to land in about thirty seconds, which is why a cold open beats a prologue: drop the player into a situation that demands action and fill in the backstory through play.
Episodic structure follows from the same constraint. Self contained episodes with their own arc, connected by a throughline, map naturally onto how people actually use a browser.
Takeaway
Spend the budget on reaction, not on routes. The full narrative design guide covers environmental storytelling, dialogue economy on small screens, and what LLM driven NPCs change about the designer's job.
