framework

Evaluate designs

System Usability Scale

A 10-question survey that scores any product's usability from 0 to 100.

⏱ 9 minBeginnerBy John BrookeSince 1986

Use it when

Measuring perceived usability and comparing versions.

You'll walk away with

A SUS score and grade per version.

Step 1 · The idea

Ten questions, one usability score

The System Usability Scale (SUS) was created by John Brooke at Digital Equipment Corporation in 1986 as a “quick and dirty” usability survey. It's now one of the most used measures of perceived usability in the world.

Users answer 10 statements on a 1–5 scale after trying a product. The result is a score from 0 to 100 that you can compare over time and against other products. The average is about 68.

SUS tells you how usable people feel something is, not why. Pair it with usability test observations.

The model

How scoring works

Alternating statements

Odd statements are positive (“easy to use”). Even statements are negative (“unnecessarily complex”). This stops people from agreeing with everything.

Convert each answer

Odd items: answer − 1. Even items: 5 − answer. Each item now scores 0–4.

Example: Answer 4 on item 3 → 3. Answer 2 on item 2 → 3.

Add and multiply

Add the ten item scores (0–40) and multiply by 2.5. That gives 0–100.

Example: Total 30 × 2.5 = 75.

Interpret

68 is average. Above 80 is excellent; below 51 means serious problems. It's not a percentage.

Example: Sauro–Lewis curved grades: A ≥ 80.8, C = 65–71, F < 51.7.

Step 3 · Real process

Measuring a redesign with SUS

A fictional payments app team redesigns its split-bill flow and uses SUS to measure the change.

Round 1 · old flow

8 participants split a dinner bill, then fill in SUS

Average SUS: 58 · Grade D

Step 1

Baseline test

First, measure the current design so there's something to compare against.

Baseline

Observations

6 of 8 couldn't tell who had already paid

4 of 8 typed the amounts twice

Step 2

Find out why

The score says “below average”. The observations say why. You need both.

Qualitative

Changes

Paid / pending badges

Saved friend groups

Clear payment errors

Step 3

Redesign

Fix the biggest problems found in testing.

Design

Round 2 · new flow

8 new participants, same tasks

Average SUS: 76 · Grade B

Step 4

Test again

Same tasks, same method, new people. The score moved from below average to good.

Compare

Fictional example; scores are illustrative.

Watch out

Common mistakes

✓ Do

Give SUS right after people use the product
Keep the standard wording (small tweaks like “website” for “system” are fine)
Compare scores for the same tasks over time
Pair it with observations and task success
Report the average and the number of participants

✗ Avoid

Reading the score as a percentage
Changing or dropping statements
Using it to find specific problems — it can't
Comparing across very different tasks or products without context
Calculating item averages instead of per-person scores first

Cheat sheet

SUS in one screen

10 STATEMENTS, answered 1 (strongly disagree) – 5 (strongly agree)
odd = positive · even = negative

SCORING (per participant)
  odd items:  answer − 1
  even items: 5 − answer
  SUS = (sum of 10 items) × 2.5        → 0 to 100

INTERPRET
  68 ≈ average · 80.8+ = A · 71–80 = B/C+ · below 51.7 = F
  not a percentage; percentile-like

USE  after tasks · same tasks when comparing · pair with observations

Use it at work

Figma template

A ready-made canvas for running this framework with your team. Free, editable, and it works in Figma and FigJam.

Figma & FigJam · free