The invisible cues that tell users what your interface will let them do

The term "affordance" was coined by the psychologist James J. Gibson in 1979 to describe what an environment offers an organism — the action possibilities it presents. A flat surface affords walking. A branch affords grasping. Gibson was interested in perception, not design.

Don Norman borrowed it for The Design of Everyday Things and applied it to designed objects. A door with a flat plate affords pushing. A door with a handle affords pulling. When the physical cue contradicts the intended action — a handle on a door you need to push — confusion follows. Norman used this to explain why some things are intuitively usable and others require instruction.

In digital products, the same logic holds. A button with a raised appearance and shadow affords clicking. A slider thumb affords dragging. A field with a blinking cursor affords typing. When these cues are absent or misleading, users get stuck — not because they're confused, but because the interface didn't communicate what was possible.

Here's where most designers get this slightly wrong. In later editions of his book, Norman clarified that what designers actually control is not affordances but signifiers — the perceptual cues that communicate what an affordance is.

The affordance (the actual possibility of interaction) exists whether or not the user perceives it. A button can be clicked even if it doesn't look clickable. The signifier is what makes that possibility visible: the shape, the shadow, the label, the hover state.

So when a button doesn't look clickable, you don't have an affordance problem — you have a signifier problem. That distinction matters for diagnosis. Affordance problems require restructuring the interaction itself. Signifier problems are visual and communicative — solvable through design.

Most teams use "affordance" colloquially to mean both. That's fine in conversation. But when you're tracking down a usability problem, the distinction helps you find the right lever.

The most recurring signifier failures in complex B2B products:

• Ghost buttons on light backgrounds: Low contrast strips away the "this is interactive" signal. Users scan past them.
• Flat cards that should be clickable: Without a border treatment or hover state, there's no cue that the whole card is a link.
• Drag-and-drop with no handle: Users don't try to drag things that don't look draggable. If the affordance exists but the signifier is missing, the feature effectively doesn't exist for most users.
• Inline editing hidden behind hover: Users don't discover it until they stumble on it accidentally — and on mobile, where hover doesn't exist, they don't discover it at all.
• Icon-only action bars: In data-dense dashboards, most users don't know what unlabelled icons mean. NNGroup's research on icon usability consistently shows labels improve task success rates, even for icons teams assume are universally understood.

Each of these causes friction in the moment and erodes user confidence over time.

A practical audit doesn't require a full research study. Two approaches that work:

First-use observation: Watch someone who's never used the product try to complete a core task. Note every pause, every hover without clicking, every missed interactive element. You're looking for places where the signifier failed — where an interaction was possible but not perceived.

Label removal test: Take screenshots of your UI and mentally — or literally — remove all text labels and colour. What's left? If the remaining shapes, shadows, and spatial relationships don't communicate what's interactive, your affordances are being carried entirely by text. That works when users are reading carefully; it breaks down when they're scanning.

Pair this with a Heuristic Evaluation using Nielsen's visibility of system status principle. Affordances are part of what makes a system's capabilities legible before the user commits to an action.

Poor affordances compound in ways that are hard to attribute directly. One missed signifier causes one missed action. But across a product with dozens of interaction points, consistently weak cues create an experience where users feel unsure of what to do next — and that uncertainty steadily increases Cognitive Load across every task.

In consumer apps, frustrated users leave. In enterprise SaaS, they can't — they're required to use the tool. So the frustration accumulates. It shows up in support volume, training costs, low feature adoption, and eventually in churn surveys as "the product is hard to use."

For teams preparing for an UX Audit, affordance quality is one of the first things a structured heuristic review covers — because it affects every screen, every task, and every user in the product.