Designing hint systems that don't spoil the puzzle

What a hint is actually for

A hint exists to rescue a player from being stuck, not from having to think. Those sound similar but they're different problems: "stuck" usually means the player is missing one piece of information or hasn't noticed one part of the board; "having to think" is the entire point of a puzzle game. A hint system that treats every pause as a problem to be solved will end up quietly solving the game for the player, one nudge at a time.

The useful question isn't "should this game have hints" — it almost always should — it's "what is the smallest true thing we can tell this player that gets them unstuck without telling them the answer."

Layered or opt-in hints vs. auto-solve

The most respectful hint systems are layered and opt-in: a first hint that only points at where to look, a second that names what to consider, and only a final, explicitly-requested step that gets closer to the answer — each one a separate, deliberate choice by the player, not something that fires automatically after a timer. Auto-solve hints, the kind that eventually just complete the puzzle if you wait or fail enough times, remove the player's agency entirely; they turn a puzzle into a countdown to being shown the solution.

Opt-in layering keeps the decision to ask for help where it belongs — with the player who knows whether they're still enjoying the struggle or genuinely done with it.

Protecting the aha-moment

The aha-moment — the instant a player's own reasoning clicks into place — is the actual product a puzzle game is selling. Every hint spends a small amount of that moment's value to buy the player forward progress. A well-designed hint spends as little of it as possible: a nudge toward a direction still leaves the click of understanding to the player. A hint that states the solution outright spends the whole thing, and no player remembers a puzzle they were told the answer to.

This is also why "no hints at all" isn't automatically the more respectful design. A player stuck for the wrong reason — a misread rule, an overlooked tile — isn't experiencing productive difficulty, they're experiencing friction, and friction doesn't protect anything worth protecting.

Where we land on this for our own game

The first Solobit Games title is still in development, so this describes design intent, not a shipped, tuned hint system. The direction we're building toward is layered and opt-in — small, player-requested nudges rather than a timer-based auto-solve — because it keeps the aha-moment intact for players who want to earn it, while still giving an honest way forward to anyone who's genuinely stuck. Whether the exact layering is right will keep getting checked against real playtesters, the same way every mechanic on this site's other notes gets checked.