Fake difficulty and real difficulty in playable design
Two playables can look equally hard and convert completely differently. The difference is whether the challenge sits in the mechanic or in the interface.
A player who fails a playable three times in a row will draw one of two conclusions. Either the game is hard and they want to beat it, or the interface is fighting them and they want to leave. Both look identical in a session recording: repeated failed attempts, rising time-per-attempt, an eventual tap on the end card or an exit. Only one of them is doing the job the playable was built for.
Real difficulty comes from the mechanic itself: a puzzle that requires genuine spatial reasoning, a timing window that is tight but learnable, a resource constraint that forces a real choice. A player who fails against real difficulty usually fails a little differently each time, because they are testing different approaches. Fake difficulty comes from friction the player did not sign up for: a hit box that does not match the sprite, a drag gesture that is difficult to register on a touchscreen at playable size, a tutorial prompt that disappears before it is read, an object that blocks the one interactive element on screen. A player who fails against fake difficulty usually fails the same way each time, because the obstacle is not the puzzle, it is the controls.
Why this matters more in a playable than in the full game
A player has roughly thirty seconds and no investment in your game yet. In the shipped app, a control quirk is an annoyance a returning player learns to work around, because they already want to be there. In a playable, the same quirk reads as evidence the game will not respect their time, and there is no accumulated goodwill to absorb it. This is also why a playable can fail to convert even when session length and completion rate look reasonable in aggregate. A completion driven by trial and error against a bad hit box does not tell a player anything true about whether they will enjoy the shipped mechanic. It just tells them the demo was fiddly and they got through it anyway.
How to tell them apart in review
Watch failed attempts, not just completion rate. If failure patterns vary between attempts, the challenge is probably real. If every failed attempt looks nearly identical, in timing, in the point of failure, in the player's apparent confusion, that is a strong signal the difficulty is sitting in the interface rather than the mechanic. A second, cheaper check: have someone unfamiliar with the build play it once, without instruction, and watch where their first failure happens. If it happens at a moment the design intended to be a genuine test, the difficulty is doing its job. If it happens at a moment nobody flagged as hard, that is friction, not challenge.
The fix is rarely more tutorial
The instinct when a playable's early drop-off looks like difficulty is to add a clearer instruction or a bigger arrow. That treats the symptom. If the underlying problem is a hit box, a gesture threshold, or an obscured tap target, more instruction just tells the player more clearly that the game is about to be annoying. Fix the interaction first. Add guidance only for challenge that is genuinely there to be understood, not for friction that should not have shipped in the build at all.
