The design of everyday things: why doors, taps and light switches confuse us

Stop 01A book about blaming the object, not the person

Norman’s book first came out in 1988 under the title The Psychology of Everyday Things, and a revised and expanded edition followed in 2013. Nielsen Norman Group’s description of it opens with the experience everybody knows: being unable to work out which light switch or stove burner to turn on, or “whether to push, pull, or slide a door.” Its argument is that the fault lies not with us but with “product design that ignores the needs of users.”

The door has become the book’s mascot. In design circles a door that makes you pull when you should push is now called a Norman door, according to the Interaction Design Foundation, because that scenario opens the book. Three ideas from it explain most everyday confusion: affordances, signifiers and mapping.

Stop 02Affordances: what an object lets you do

An affordance is a possible action. A door affords opening. A chair affords sitting. A button affords pressing. Norman borrowed the word from the perceptual psychologist J. J. Gibson, who used it for the relationship between an object and whoever is using it. As Norman notes on his own site, for Gibson affordances “did not have to be perceivable or even knowable — they simply existed.”

That last point is the problem. A door can be perfectly able to swing both ways and give you no clue which way it does. The possibility exists. You just cannot see it.

Stop 03Signifiers: what tells you what to do

Norman added a second word to fix this. In a 2008 column in the ACM magazine Interactions, later posted on his site as “Signifiers, not affordances,” he defines a signifier as “some sort of indicator, some signal in the physical or social world that can be interpreted meaningfully.” The 2013 edition of the book builds the idea in.

On a door, the classic signifiers are the hardware itself. Early editions of the book, the Interaction Design Foundation notes, described “flush plates for pushing and protruding handles for pulling.” A flat plate gives your hand nothing to grip, so you push. A bar or loop invites a grip, so you pull. When a door opens only one way but has the same pull handle on both sides (common, the same article says, because designers like symmetry), one side is lying to you.

The “PUSH” sticker taped to the glass is also a signifier. It is a patch. A well-designed door would not need it.

Norman’s column is blunt about which idea matters more for anyone who makes things: “Designers of the world: Forget affordances provide signifiers.” He also points out that signifiers can be social. The worn path across a park lawn, which planners call a desire line, tells you where people actually walk. It was never designed at all.

Stop 04Mapping: which control does what

The third idea explains the stove and the light switch. Mapping is the relationship between controls and what they control. Nielsen Norman Group defines natural mapping as a design “in which the system’s controls represent or correspond to the desired outcome,” and notes that such systems “are faster to learn and easier to remember.”

The stove is the textbook case. Four burners sit in a two-by-two square, but the four knobs sit in a single row along the front. Which knob is the back left? You read the tiny diagram or you guess, and you guess wrong often enough to learn to hover your hand over the burner. NN/g contrasts this with a stove whose knobs are laid out in the same pattern as its burners. The idea has deep roots: the article traces it to 1950s experiments by Paul Fitts and Charles Seeger on stimulus-response compatibility, published in 1953.

A bank of identical light switches by a door has the same problem. Three switches in a row, three lights scattered across a room, and no relationship between them. Most people end up flicking all three.

Stop 05Why taps confuse us

A kitchen or bathroom tap fails on all three counts at once, which is why hotel showers are a small daily puzzle.

• Affordance without a signifier. A smooth lever might turn, slide, push or pull. Nothing tells you which.
• Weak mapping. Hot and cold are often marked with small red and blue dots that are worn off, hidden or missing, and a single lever may give you hot on the left, on the right, or at the back.
• Slow feedback. Among the principles Wikipedia summarizes from the book is feedback: “the user receives full and continuous feedback about the results of the actions.” Water temperature takes seconds to change, so you overcorrect, then correct back.

None of this makes you bad at taps. It makes the tap bad at telling you.

Stop 06What good design does instead

Nielsen Norman Group sums up the book’s rules in three phrases: “make things visible,” “exploit natural relationships that couple function and control,” and “make intelligent use of constraints.” Once you know them, you see them everywhere:

• A door with a flat plate on the push side and a vertical handle on the pull side. No sign needed.
• Stove knobs laid out in the same square as the burners they control. No diagram needed.
• A switch plate installed so that the switch for the front light is at the front and the one for the back light is at the back.
• A plug that only fits one way. That is a constraint doing the work a warning label would otherwise do.

Stop 07Using it in your own work

You do not need to design doors for this to matter. Every form, invoice, menu, booking page and sign you make is an everyday thing to somebody else. A few questions borrowed from Norman help:

1. 01Can someone tell what to do just by looking? If you need a note explaining how to use it, the design is doing too little.
2. 02Do the controls sit where their effects are? Put the “Pay” button next to the total, not at the top of the page.
3. 03Does the person get feedback right away? A confirmation message, a sound, a change of colour.
4. 04Can they make the wrong move at all? If a mistake is possible and costly, use a constraint to make it impossible.

The next time you pull a push door, you can skip the embarrassment. Look at the handle instead and ask who put it there.

I’m Saliem Talash. I edit ZELR, a publication on tools, work and the city, and this is the lens I use for all of it.