Fitts's Law

Law 15 / 20 Design Better Decisions

Targets that are larger and closer are faster and more reliable to hit, but making everything big quickly turns into clutter and mistaken taps.

  • UX model or heuristic
  • 5 min read
  • Documented study
  • Reviewed by Ali Dezhami, November 8, 2026

The dilemma

Illustrative scenario

Priya leads digital product at a regional animal-rescue charity. Mobile visitors now make up most donation traffic, and session recordings show the same pattern again and again: people choose an amount, then hunt for a small "Continue" text link squeezed between "Edit amount" and "Cancel". Some tap Cancel by accident and leave.

The design agency proposes a full-width sticky donate button pinned to the bottom of every page. Mobile completion looks likely to rise. But the fundraising director objects: the same agency's earlier redesign raised clicks on a newsletter banner while sign-up quality fell, and she worries that a giant button on every page will pull people into the form before they have read why the charity needs the money, inflating starts and refunds.

Priya has to decide how big, how close and how often the main action should appear, and what to measure so that easier tapping does not hide worse decisions.

The short answer

Fitts's Law is a model of aimed movement: the time it takes to point at a target rises with the distance to the target and falls as the target gets wider. Psychologist Paul Fitts described it in 1954, summarizing the difficulty of a movement as an index that grows with the ratio of distance to target width. In interface terms, the next important action should be easy to reach and large enough to hit without precision.

How strong is the evidence?

UX model or heuristic

A model or design heuristic from human-computer interaction. Useful for predictions, not a guarantee of conversion.

This is a UX model built on a well-replicated motor-control finding. Fitts measured people tapping between targets, and moving discs and pins, at varied distances and widths; movement time rose in a predictable way as targets got farther away or narrower. Human-computer interaction researchers later found the same relationship when people point with a mouse and other devices, and refined the formula.

Three conditions matter for marketers. First, the model predicts movement time and error rates, not whether someone wants to click; a button that is easy to hit does not create intent. Second, it was built for continuous pointing. On touchscreens the finger covers the target and has its own precision limit, so spacing and minimum size matter as much as the formula. Third, accessibility guidance sets a floor: WCAG 2.2 success criterion 2.5.8 (level AA) asks for pointer targets of at least 24 by 24 CSS pixels, or enough spacing around smaller ones, with exceptions such as links inside a sentence.

Investigating an example

Documented study

The original pointing experiments

In his 1954 paper, Fitts asked participants to tap alternately between two metal strips with a stylus, and to transfer discs and pins between holders, while he varied the distance between targets and their width. Movement time increased steadily as the index of difficulty rose, which let him describe aimed movement with a single relationship between distance and width.

Limits: these were laboratory tasks with practiced participants, a stylus and one-dimensional movement. The model says nothing about attention, motivation or whether a person should click at all. Touch input, two-dimensional layouts and users with tremor or low vision need their own testing.

A published study or record, linked below. Its limits are stated.

Choose your move

You are in the dilemma above. Which move would you make? Open each one to see its trade-offs.

A Ship the full-width sticky donate button on every page.

Tapping becomes easy and form starts will likely rise, but the button competes with the story on every screen and can pull in people who are not ready. Watch completed gifts and refunds, not starts.

B Make Continue a clearly labeled primary button in the thumb zone of the donation step, and move Cancel away from it.

Fixes the real error at the moment of decision without adding pressure elsewhere. Gains may be smaller in the headline numbers, and it needs testing on several phone sizes.

C Keep the layout and add a tooltip explaining where Continue is.

Cheap and reversible, but it adds reading to a motor problem. People who mis-tap Cancel still leave before they ever see the tooltip.

No move wins in every business. The law tells you what to check, not what to choose.

How to apply it

  1. Find the high-value targets. List the two or three actions that finish a task (continue, add to cart, pay) and the destructive ones next to them (cancel, remove).
  2. Measure size and spacing. Check each target against the 24 by 24 CSS pixel WCAG minimum or its spacing exception, on the smallest phone you support.
  3. Reduce distance where attention already is. Place the next step close to the last thing the user touched, and keep destructive actions farther away or behind a confirmation.
  4. Test against the current layout. Run an A/B test or a moderated session comparing mis-taps, task time and completed tasks, not just clicks on the bigger button.
  5. Check the downstream quality. Compare refunds, cancellations or support contacts between versions to make sure easier tapping did not bring in hurried decisions.

What to measure

Mis-tap rate (taps on the wrong control or on nothing), time from previous step to the next tap, completion rate of the whole task, and back-button or cancel rate. Guardrails: refund or cancellation rate, and accessibility checks for target size and focus order.

Worked example

Worked example: two layouts for the same mobile donation step (illustrative numbers)
MeasureSmall text linkLarge primary button
Visitors reaching the step 2,000 2,000
Accidental taps on Cancel 140 30
Completed donations 760 840
Refund requests within 30 days 12 14

Illustrative scenario, not data from a real charity. Assumes visitors were randomly split and the copy was identical. The point: the bigger target mostly removes accidental exits; the refund row is the guardrail that tells you whether easier tapping changed decision quality.

Where it breaks

  • Bigger is not always better. Enlarging every control flattens the visual hierarchy, so nothing stands out and the screen gets crowded.
  • The model predicts speed, not desire. A large button cannot rescue a weak offer or an unclear price.
  • Touch has its own limits. Fingers hide what they press; spacing between neighbors often matters more than raw size.
  • Proximity cuts both ways. Putting Delete or Cancel next to the primary action makes costly mistakes easy too.
  • Ethics: making Accept huge and Decline tiny uses the same physics to steer people against their interest. It is widely criticized as a dark pattern.

What if you apply it incorrectly?

A retailer reads Fitts's Law as "make the buy button as big as possible" and pins an oversized Buy now bar over product details on mobile. Add-to-cart taps jump. Two weeks later, returns climb because shoppers skipped the size chart the bar was covering, and accessibility complaints arrive because the bar hides content when the page is zoomed.

Myth versus evidence

The social-video version

Make the call-to-action button as big as you can and conversions will go up.

What the evidence supports

Fitts's Law predicts how quickly and accurately a target can be hit; size helps up to a point, but placement, spacing and the offer decide whether people convert.

Five-minute field assignment

Open your checkout or sign-up flow on the smallest phone you can find and complete it with one thumb. Note every near-miss, every target you had to zoom to hit, and every place where a destructive action sits next to the primary one. Then check your analytics for the step with the highest exit rate: is it one of those screens?

Checkpoint

Two questions: one about the idea, one about a decision. Answer both correctly to master this law. Questions are drawn from a larger bank and graded on our server; a retry brings different questions.

Evidence ledger

Every factual claim in this lesson, where it comes from, and how far it has been checked.

Further reading

Which other law could explain this decision?

Editorial review: Ali Dezhami, November 8, 2026. Last substantive update: November 8, 2026. Our methodology