What Is Software Testing and Why Does It Matter in 2026?

FEATURE: Age field on a sign-up form (allowed 18 to 60)

Positive tests
  - Enter 25          -> accepted

Negative tests
  - Enter 17          -> error: must be 18 or older
  - Enter 61          -> error: must be 60 or under
  - Enter abc         -> error: numbers only
  - Leave blank       -> error: required field
  - Enter -5          -> error
  - Enter 25.5        -> error or rounded? (ask the analyst)

The last question is a requirement gap - testers find those too.

BOUNDARY VALUE ANALYSIS (range 18-60)

Value   Expected     Why test it
-----   ----------   --------------------------
17      Rejected     Just below the lower limit
18      Accepted     Exactly on the lower limit
19      Accepted     Just above the lower limit
59      Accepted     Just below the upper limit
60      Accepted     Exactly on the upper limit
61      Rejected     Just above the upper limit

Defects cluster at edges, so test both sides of each limit.

// login.spec.ts - Playwright Test (npm init playwright@latest)
import { test, expect } from '@playwright/test';

test('shows error for wrong password', async ({ page }) => {
  await page.goto('https://example.com/login');
  await page.getByLabel('Email').fill('user@example.com');
  await page.getByLabel('Password').fill('wrong-password');
  await page.getByRole('button', { name: 'Sign in' }).click();
  await expect(page.getByText('Invalid email or password')).toBeVisible();
});

// Run with: npx playwright test

# Check that an API returns 200 and JSON (replace the URL)
curl -s -o response.json -w "%{http_code}\n" \
  https://api.example.com/v1/products/101
# -> 200

# Negative test: a product that does not exist
curl -s -o /dev/null -w "%{http_code}\n" \
  https://api.example.com/v1/products/999999
# -> expected 404, not 500

# A 500 for a missing item is a bug: the API should
# handle 'not found' cleanly.

# .github/workflows/tests.yml - run tests on every push
name: tests
on: [push, pull_request]
jobs:
  e2e:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with:
          node-version: 22
      - run: npm ci
      - run: npx playwright install --with-deps
      - run: npx playwright test

TESTER'S CHECKLIST FOR AI-ASSISTED CODE

Feature: apply discount coupon

[ ] Expired coupon rejected?
[ ] Coupon applied twice -> only once?
[ ] Discount larger than order total -> total not negative?
[ ] Coupon for another user's account blocked?
[ ] Case sensitivity: SAVE10 vs save10 (check the rule)
[ ] Error messages clear and not leaking internals
[ ] Unit tests exist for each rule above

AI tools often write the happy path well and miss these.

TEST CASE SHEET

ID     Title                         Pre-condition     Steps                         Test data          Expected                 Status
-----  ----------------------------  ----------------  ----------------------------  -----------------  -----------------------  ------
TC-01  Login with valid details      User registered   1 Open login 2 Enter 3 Submit  user@example.com   Dashboard opens          Pass
TC-02  Login with wrong password     User registered   Same as TC-01                 wrong password     Error message shown      Pass
TC-03  Lockout after 5 failures      User registered   Enter wrong password 5 times  wrong password x5  Account locked message   Fail
TC-04  Password field hides input    None              Type in password field        any text           Characters masked        Pass

BUG REPORT

ID          : BUG-214
Title       : Account not locked after 5 wrong passwords
Environment : QA, Chrome on Windows 11, build 3.8.2
Severity    : High (security)     Priority: High
Steps       :
  1. Open the login page
  2. Enter a registered email
  3. Enter a wrong password 5 times
Expected    : Account locked message after 5th attempt
Actual      : 6th and 7th attempts still allowed
Evidence    : Screen recording, network log attached
Linked test : TC-03

RELEASE TESTING CHECKLIST

Build / version: ________   Date: ________

[ ] Smoke test passed on the new build
[ ] All new feature test cases executed
[ ] Regression suite run (manual + automated)
[ ] Critical and high bugs fixed and retested
[ ] Open bugs reviewed and accepted by product owner
[ ] API checks for changed endpoints passed
[ ] Basic performance check on key pages
[ ] Test summary report shared with the team

Decision: Go / No-go   Signed off by: ________