---
name: Proximity
slug: proximity
category: layout
status: published
created: 2026-08-21T00:00:00.000Z
modified: 2026-08-21T00:00:00.000Z
definition: The grouping principle that elements placed close together read as
  related, which is why spacing decides perceived structure more reliably than
  borders do.
aliases:
  - name: law of proximity
    source: community
  - name: Gestalt proximity
    source: nngroup
  - name: grouping by proximity
    source: community
tags:
  - perception
  - spacing
relations:
  contrastWith:
    - common-region
    - whitespace
  variantOf: []
  partOf: []
  seeAlso:
    - chunking
implementations: []
sources:
  - title: Proximity Principle in Visual Design, NN/g
    url: https://www.nngroup.com/articles/gestalt-proximity/
demo: inline
exhibit: false
useWhen: arguing that spacing, not lines, makes groups
---

Proximity is the oldest of the Gestalt grouping principles, described by Max Wertheimer in
1923 and rediscovered by every generation of interface designers since. The claim is
narrow and strong: things placed near each other are perceived as one thing, and the
perception happens before reading, not after. That is what makes it useful. A reader has
already decided which label belongs to which field, and which price belongs to which
product, in the time it takes to notice that the screen contains a form at all. Every
other grouping device is competing with a judgement that has already been made.

The working rule is a ratio rather than a number. Space inside a group has to be visibly
smaller than the space around it, roughly half or less, and if the two are within a step of
each other on your spacing scale the grouping simply does not read. This is also the answer
to the most common review note in interface design, which is that a screen feels
disorganised. It usually is not missing dividers. It is spending the same gap everywhere,
so the structure has to be reconstructed from the words. [Whitespace](/whitespace) is the
material this principle spends, and a spacing scale with widely separated steps is what
makes spending it possible; a scale of near neighbours can only produce walls.

The classic failure is the form label. A label placed midway between the field above it and
the field below it belongs to neither, and the reader resolves the ambiguity by guessing,
usually correctly and occasionally not. Nielsen Norman Group's
[write-up of the principle](https://www.nngroup.com/articles/gestalt-proximity/) has the
same example, because it keeps happening: the fix is to move the label three pixels closer
to its own field and add twenty between the pairs, which costs nothing and removes the
question. The same arithmetic settles captions under images, helper text under inputs, and
prices under product names.

Proximity has neighbours worth naming. Common region groups by enclosure, drawing a border
or a shared background around the members, and it wins when items must be grouped without
being moved, as in a settings page whose sections cannot be reordered. Similarity groups by
shared appearance, which is what a colour-coded legend relies on. All three can say the same
thing, and saying it three times is how a page ends up loud. Proximity is the one to reach
for first because it is silent: it adds no ink, it survives a theme change, and it does not
compete with the content for attention. The place it loses is density. A table of two
hundred rows has no room to spend, so the honest tool there is a rule, not a gap.