cognitive psychology
Gulf of execution and evaluation
What are the gulfs of execution and evaluation?
Two gaps sit between a person and a system. The gulf of execution is the distance between what you want to do and figuring out how to do it - "I know my goal, but how do I make this thing do it?". The gulf of evaluation is the distance between the system acting and you understanding what happened - "I did something, but did it work?". Good design narrows both; bad design leaves you stranded on either side.
Also known as: gulf of execution, gulf of evaluation, gulfs of execution and evaluation, norman gulfs
Prefer to watch?
Watch the recap 0:49
The demo
Your goal: save your work. Flip the gulfs wide, then narrow, and feel the two separate frustrations - not knowing how to act, then not knowing whether it worked.
What this demo shows (text version)
The task is simply to save your work. With the gulfs wide, the execution gulf is open: the only control is an unlabelled mystery icon that gives no clue it saves, so you must guess how to act. And the evaluation gulf is open too: after you press it, nothing visibly changes, so you can't tell whether it worked. With the gulfs narrow, a clearly labelled "Save" button makes the action obvious (execution bridged), and an instant "Saved ✓" confirmation makes the result obvious (evaluation bridged).
The two gulfs are the spine of how we interact with anything: first work out how to act, then work out whether it worked. Execution is bridged by affordances, signifiers and visible options; evaluation by clear, immediate feedback. Almost every usability problem lives in one gulf or the other, and naming which one tells you the fix.
Every interaction crosses two gulfs: working out how to act (execution) and working out whether it worked (evaluation). Bridge execution with clear affordances, signifiers and visible options so the right action is obvious; bridge evaluation with immediate, legible feedback so the result is obvious. A confusing interface is one that makes you guess how to act, then guess whether you succeeded.
With the gulfs wide, you were stuck twice over - unsure how to do the thing, then unsure if it had worked. Narrow them and the friction vanishes: the control says "press me", and the feedback says "done". Same task; the difference is whether the design bridges the gaps or leaves you stranded on both sides of them.
The gulf of execution is about acting: can you tell what's possible and how to do it? It's bridged by affordances and signifiers (the thing looks and labels itself as doable), by making the options visible rather than hidden, and by matching the controls to your intentions so there's a clear path from goal to action. A wide execution gulf feels like "I have no idea how to make this happen".
The gulf of evaluation is about understanding: once you act, can you tell what the system did and whether it matched your goal? It's bridged by feedback - immediate, perceivable, interpretable - that reports the new state clearly. A wide evaluation gulf feels like "I pressed something… did anything happen?", the dread of the unresponsive button and the silent failure.
Together they're the spine of Norman's model of interaction (and his seven stages of action sit inside them). Almost every usability problem lives in one gulf or the other: people can't work out how to act, or can't tell whether their action worked. Diagnosing which gulf a problem sits in tells you the fix - clearer affordances and signifiers, or clearer feedback.