The smallest change a person can perceive is always proportional to what they are already experiencing — not fixed. A 1kg change is obvious when you're holding 2kg and invisible when you're holding 50kg. In design: the same absolute change in type size, spacing, or price produces completely different felt effects depending on the starting point.
In 1834, German anatomist Ernst Heinrich Weber discovered that the threshold for detecting a difference between two stimuli is not a fixed quantity — it is a ratio. He was studying weights: what is the smallest difference between two objects that a person can reliably feel? His answer was not “1 gram” or “5 grams.” It was approximately 1/40th of the original weight. Holding 40g, you can just detect 1g of difference. Holding 400g, you need 10g of difference to notice the same change. The ratio stays constant. The absolute number scales with the stimulus.
Gustav Fechner formalised this into Weber's Law: the just noticeable difference (JND) equals a constant fraction of the original intensity. For visual stimuli — size, brightness, spacing — the JND is typically somewhere between 10% and 30%. Below that threshold, changes exist but are not perceived. Above it, they register as intentional and meaningful.
The design implication is direct and frequently violated. A 2px size increase on a 12px label is a 17% change — perceivable. The same 2px increase on a 48px heading is a 4% change — invisible. A $20 discount on a $49 plan is 41% — felt as a real deal. The same $20 on a $399 plan is 5% — below the threshold of felt significance.
“The threshold for noticing a change is not a fixed amount — it is a fixed fraction of what already exists.”
The most common violation of Weber's Law in product design is a type scale where the size differences between heading levels are below the JND. On paper — or in Figma at 100% zoom — a move from 20px to 24px looks meaningful. In a browser, at normal reading distance, it reads as an inconsistency rather than deliberate hierarchy.
Gaps between H2 (26px) and H3 (22px) and H4 (18px) are 15-18%. Below the JND at these sizes. Four levels that look almost the same.
Every step is 33% larger. H2 (31px) vs H3 (23px) vs H4 (17px) — each jump is clearly perceivable at any screen size.
The proportional scale looks more extreme — particularly the large 41px display size. That is exactly the point. Weber's Law requires the top of the scale to be significantly larger than designers trained on linear systems expect, because the eye needs a proportionally large difference to register a level of hierarchy as distinct.
Spacing is where Weber's Law fails most silently. A designer carefully specifies 8px, 12px, and 16px spacing for different levels of visual hierarchy in a component — and in the browser, users perceive three identical gaps. The 4px differences are below the JND for spacing at these values.
Header padding (12px) vs body (12px) vs footer (8px). The 4px difference is below the JND. All three sections feel equally weighted.
Header (10px), body (20px), footer (8px). Each zone is perceivably different — the content area has breathing room the eye registers as primary.
Price perception follows Weber's Law precisely. A $20 discount on a $49 product is a 41% saving — strongly felt. The same $20 on a $399 product is 5% — below most users' threshold of significance. Anchoring a high-price tier next to a target tier exploits the proportional nature of value perception: the Enterprise plan's job is often perception, not conversion.
$49/mo feels expensive in isolation. No reference point.
$49/mo looks proportionally affordable next to $199/mo. Same plan, different felt value.
Weber's Law is not a design heuristic — it is a psychophysical constant. The JND for visual design is approximately 10–30% depending on the dimension. Any intentional difference below that threshold is perceptually wasted. Any unintentional difference above it looks like a mistake. The law gives designers a precise tool for auditing their own decisions: check every delta as a percentage.
Weber, E. H. (1834). De Pulsu, Resorptione, Auditu et Tactu. Leipzig: Koehler. Fechner, G. T. (1860). Elemente der Psychophysik. Stevens, S. S. (1957). On the psychophysical law. Psychological Review, 64(3), 153-181.