Reducing the reachability of dangerous actions

What it is

Reducing reachability means increasing the path length, visibility, or number of steps required to reach a high-consequence action, so it's harder to trigger by accident, in passing, or in a hurry — without making the action impossible to carry out. This is different from blocking an action outright: an outright block means the action can't be completed until some precondition is met, whereas this is about burying the action deeper, moving it farther away, or requiring more steps or a more deliberate trigger — the action can still be completed whenever the user genuinely wants it.

Why it happens

Accidental and casual misoperations mostly happen when an action sits close to a well-worn path and costs almost nothing extra to hit — a finger sliding across the screen, a glance passing over it, a habitual run of taps can all land on it in passing. Moving a dangerous action off the common path, adding an extra level of navigation, or requiring a higher-precision trigger like a long press effectively adds an "intent confirmation cost" to it: an accidental, unconscious motion rarely clears several layers of friction in a row, while a user who genuinely wants to perform the action is willing to pay those extra steps. What friction raises is the threshold for unintentional triggering, not the threshold for deliberate execution — and that's exactly how it divides labor with other error-proofing devices like defaults or input-time constraints.

Where it stops holding

Reduced reachability doesn't work for high-frequency actions — if users need to perform the action often, the extra path and steps become a standing efficiency tax, and they'll quickly work out a fixed routine to route around the friction, at which point the friction is friction in name only. It also can't substitute for the hard block an irreversible or high-consequence action needs: for an action that can't be undone once triggered, merely being harder to reach isn't enough to carry the risk, because once someone actually reaches it, the action still executes in full.

Applying it

Identify actions in the interface with asymmetric consequences — where the cost of one accidental hit far exceeds any benefit — and move them off the physical or visual path of routine actions: put them in a secondary menu, require expanding a section first, or switch the trigger from a tap to a long press. Never place such actions next to high-frequency routine actions or bind them to the same gesture. Verification: track the rate at which the action is immediately undone or canceled right after being triggered — a high rate signals most triggers are accidental rather than intended. If that rate drops noticeably after reducing reachability, the friction is placed correctly; if it doesn't move, the problem is more likely the action's distinctiveness than its reachability.

Related

• Same group: A10.06.1 physical shape and interface constraints against mis-assembly · A10.06.3 defaults as the cheapest error-proofing device · A10.06.5 error-proofing has an efficiency cost that must be budgeted, not stacked without limit
• Nearby: A10.04 description-similarity errors · A10.14 forcing functions and interlocks
• Search terms: friction by design · hazard reachability · accidental activation