Skip to content

Stockertim Dictionary · Artificial intelligence · Technology

Pass rate

Also called pass rates, pass percentage, test pass rate

In short

A pass rate is the share of items, test cases or people that meet the standard of a test or check, worked out as the number passed divided by the number tested.

  • Beginner
  • Ratio or metric
  • Checked

In detail

The pass rate is one of the simplest measures of quality. Count how many items were tested, count how many met the standard, and divide. In testing automations and AI outputs, the items are test cases, the standard is an agreed rubric or set of expected outputs, and the result shows how often the system gets things right on cases chosen in advance.

The overall figure hides as much as it shows. A system can pass 95 per cent of routine cases and 50 per cent of difficult ones, and an overall rate of 90 per cent says nothing about which. Test sets are therefore grouped by kind of case, such as standard, unusual, ambiguous and adversarial, and a pass rate is reported for each group. Critical cases are often held to a stricter rule: a single failure on a case involving payments may block a release regardless of the total.

The number tested matters. A pass rate from 20 cases moves by 5 percentage points with each case, so small differences between versions may be noise; reporting the counts alongside the percentage lets readers judge.

The same measure appears well beyond software. Exam boards publish the share of candidates reaching a pass mark, and recruiters track the share of applicants passing each stage, where it is usually called a selection rate. In every setting the rate depends on the standard: changing the rubric or the pass mark changes the figure without anything else improving.

Formula

Pass rate = items passed / items tested x 100%

items passed
test cases, outputs or people that met the agreed standard
items tested
all test cases, outputs or people assessed against that standard

Worked example

Driftwood Travel tests a new version of its booking-confirmation automation on 60 cases.

  1. Standard cases: 38 of 40 pass, so 38 / 40 = 95%
  2. Unusual cases: 12 of 20 pass, so 12 / 20 = 60%
  3. Overall: 50 / 60 = 83.3%

The overall pass rate of 83.3% hides a 60% rate on unusual bookings, which is where the next round of work goes.

In practice

Teams report pass rates per group of test cases as well as overall, show the counts beside each percentage, set a minimum rate and a no-failure rule for critical cases before release, and compare versions on the same test set.

Watch out

  • Comparing pass rates from different test sets or different rubrics is not a fair comparison.
  • A high overall rate can conceal a poor rate on the cases that matter most.

Learn it properly

Official sources

Facts last checked on 9 November 2026.

Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.

English