Slips — correct intention, deviated execution

What it is

A slip is a failure where the user's intention and plan were correct, but the plan went wrong somewhere in the process of turning it into a concrete action — what got done is not what was planned. The test is straightforward: put the moment under retrospective scrutiny, and if the user would say "I meant to do something else, my hand/action went off," it's a slip. If what happened is actually what the user meant to happen, it isn't a slip — the problem lies elsewhere.

Why it happens

Slips occur at the level of action execution, usually inside an already well-practiced, largely automatized sequence that gets interrupted or overridden mid-run — either a stronger habit wins out at a choice point, or a momentary attentional lapse skips the check that would have confirmed the current state matches expectation. The plan itself was fine; what fails is the step that cashes the plan out into motion. This is also why slips concentrate on the most practiced, least-deliberated actions rather than unfamiliar tasks — unfamiliar actions get checked step by step and are less likely to drift off-plan.

Studying it

The standard way to separate a slip from a genuine intention is to have the user give an immediate retrospective report right after the action, then compare the stated intention against what actually happened — a match means it really was intended; a mismatch accompanied by an on-the-spot "oops, wrong thing" recognition marks it as a slip. This has to be captured right after the action: any delay lets the outcome contaminate the account, and post-hoc rationalization gets mistaken for the intention that was actually held at the time.

Where it stops holding

This test depends on users being able to accurately recall and honestly report their pre-action intention, and retrospective self-report is not always reliable — highly automatized actions often never surface a clear intention into consciousness at all, so users may not be able to say what they "meant to do" and can only infer it backward from the outcome. Also, once the same deviation recurs and the user is plainly aware of the correct action yet keeps doing it anyway, this has moved past a one-off execution slip into a deliberate, repeated choice, and slip is no longer the right label.

Applying it

When triaging an error report, ask the user directly whether what happened was what they meant to do. A "no, I fumbled/mis-tapped" response marks it as a slip candidate, and the fix belongs on the interface-constraint side — widen target spacing, change the trigger mechanism, increase visual distinctiveness between look-alikes — not user education or warning copy, since the user already knew what to do; knowledge was never the gap. Verification: filter suspected-slip error logs by whether the preceding context is highly similar and the action sits on a high-frequency path; if the same users repeatedly show the same deviation at the same step, and post-hoc interviews confirm "I knew what to do, but my hand did something else," that confirms a slip rather than another error type.

Related

• Same group: A10.01.2 Mistake — the intention itself rests on a wrong mental model · A10.01.3 Violation — deviating on purpose despite knowing the rule · A10.01.4 Slips are fixed by interface constraints, mistakes by model correction — the remedies are not interchangeable · A10.01.5 Classification must precede any redesign · A10.01.6 Blaming "user carelessness" halts diagnosis · A10.01.7 The line between violation and mistake is whether the deviation was intentional
• Nearby: A10.05 Capture errors · A10.09 Human reliability and blame culture
• Search terms: slip · action slip · error classification