Test a website by starting with the user journeys that matter, then combine functional, browser/device, performance, accessibility, and maintenance checks. No single automated scan can prove that an entire site works for every visitor. The ten practices below provide a repeatable plan for finding failures before launch and after each significant change.
1. Start with user journeys and risk
List the tasks visitors and customers must complete, then rank them by user and business impact. Typical journeys include finding product or support information, creating an account, signing in, submitting a contact form, checking out, or completing a download.
Build a risk-based test list
- Write the journey in user language, including its expected outcome.
- Identify failure consequences: lost revenue, inaccessible content, data loss, security exposure, or minor visual inconvenience.
- Test high-impact paths first and add lower-risk cases as capacity allows.
- Attach the test to a release trigger, such as every deployment or a content change.
web.dev’s test-automation guidance recommends deciding what to test and prioritizing cases instead of equating a large test count with quality.
2. Test what users can see and do
Assertions should describe rendered content and observable behavior: a heading is visible, a validation message appears, a menu opens, or a purchase confirmation is shown. Avoid coupling tests to private function names, internal data structures, or CSS classes that users never encounter.
#1 Best Overall
Prefer resilient selectors and outcomes
- Locate controls by accessible role, label, or visible text.
- Submit realistic input and verify the resulting page or message.
- Check keyboard and pointer interactions where both are supported.
- Use stable test identifiers only when a user-facing locator is not practical.
Playwright’s Best Practices makes the same point: automated tests should verify that application code works for end users and avoid implementation details.
3. Cover the browser and device matrix that matters
Use audience data from analytics, support tickets, and product requirements to choose browsers, operating systems, viewport sizes, and input methods. A site serving mostly desktop users may need a different matrix from one whose traffic is predominantly mobile. Platform and browser applicability can also differ for accessibility tools, as W3C explains in its tool-selection guidance.
Record a practical matrix
| Dimension | Examples to decide | What to verify |
|---|---|---|
| Browser | Current Chrome, Edge, Firefox, Safari; supported older versions if required | Critical journeys, layout, downloads, media, and authentication |
| Device and viewport | Representative phones, tablets, laptops, and wide monitors | Responsive breakpoints, touch targets, scrolling, orientation |
| Input | Mouse, touch, keyboard, screen reader combinations | Focus order, activation, gestures, and equivalent actions |
Do not claim that testing every possible browser is necessary; cover the combinations your audience and support policy actually require.
Rank #2
4. Keep functional tests independent
Each test should create or receive the state it needs rather than relying on another test’s order. Isolate accounts, database records, local storage, cookies, and permissions so a failure can be reproduced on its own.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesIsolation checklist
- Generate unique test data or reset fixtures between runs.
- Authenticate through a controlled setup step and avoid sharing mutable sessions.
- Clean up records that could affect later tests.
- Capture the URL, browser, viewport, input data, and application build when a test fails.
Independent tests can run in parallel and make intermittent failures easier to diagnose. Playwright documents isolation and reproducibility as core testing practices.
5. Measure page performance, then diagnose it
Use a measurement tool to establish how pages perform for real or simulated users, and a diagnostic tool to find the cause. web.dev’s performance guidance describes PageSpeed Insights as a measurement aid and Chrome DevTools as a debugging resource.
Useful performance workflow
- Measure representative landing, content, account, and transaction pages.
- Record the test conditions, including device class, connection profile, location, and whether the result is field or lab data.
- Inspect network requests, JavaScript execution, images, fonts, caching, and server timing in DevTools.
- Fix the largest causes, then measure again under the same conditions.
Treat scores as evidence for investigation, not as a guarantee of rankings, conversions, or revenue.
6. Check loading, visual stability, and responsiveness separately
Performance has distinct user experiences that should not be collapsed into one number:
Free tools Windows power users keep installed
One-click scans. No signup required.
- LCP (Largest Contentful Paint): how quickly the main visible content loads.
- CLS (Cumulative Layout Shift): whether content moves unexpectedly while loading.
- INP (Interaction to Next Paint): how promptly the page responds to user input.
Test slow and fast connections, cold and repeat loads, long pages, and interactions such as menus, search, filtering, and form submission. A page can load quickly yet respond slowly, or respond well while shifting images and buttons; investigate each symptom separately.
Rank #4
7. Run automated accessibility checks early and often
Integrate an automated accessibility checker into local development, pull requests, and scheduled scans. It can efficiently identify some missing labels, contrast problems, invalid structures, and keyboard-related issues before they reach production.
Understand automation’s boundary
Automated output is not a complete accessibility determination. Review each finding, remove false positives carefully, and track confirmed defects to closure. The W3C accessibility-evaluation overview explains why tools cover only part of an evaluation.
8. Add manual accessibility and usability review
Have knowledgeable reviewers operate the site with keyboard-only navigation, zoom or text resizing, and relevant assistive technologies. Check focus visibility and order, meaningful link and control names, error recovery, headings, instructions, dynamic updates, and whether equivalent information remains available without a particular gesture.
Include disabled users when possible
Usability testing with people who have disabilities reveals barriers that automated rules and internal reviews can miss. W3C’s Understanding Conformance guidance describes conformance as requiring automated testing combined with human evaluation; conformance and practical usability are related but not interchangeable.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.9. Choose tools for the job, not by checklist size
Before adopting a testing product or service, compare the dimensions that affect your workflow:
| Question | Why it matters |
|---|---|
| Automated checks or manual-review support? | Determines whether the tool finds rule-based defects, supports human investigation, or both. |
| Single page or site-wide scope? | Changes crawl setup, coverage, and reporting effort. |
| Browser, OS, and content support? | Ensures the tool can exercise your actual environments and formats. |
| Which standards? | Confirms that accessibility, performance, or security checks match your requirements. |
| Reporting and team workflow? | Controls triage, ownership, history, and integration with tickets or CI. |
| Free/open-source, commercial, or enterprise licensing? | Sets cost, support, hosting, and governance expectations. |
W3C’s Selecting Web Accessibility Evaluation Tools recommends evaluating scope, standards, platform support, and workflow rather than treating one tool as proof that a whole site is usable or accessible.
10. Maintain the suite and revisit coverage
Website testing is a maintenance activity. Update browser-automation dependencies, review supported browser versions, and revise journeys when navigation, content, payment, authentication, or integrations change. Playwright notes that updates allow teams to test against recent browser versions.
Recommended Free Tools
Use a recurring review
- After each release: run the critical journey smoke tests.
- On a schedule: run the full browser/device, performance, and accessibility sets.
- After incidents: add a regression test for the failure and its recovery path.
- At planning reviews: retire obsolete cases and add newly important user journeys.
Capture repeatable visual evidence
When a defect depends on layout, responsive behavior, consent dialogs, or a particular state, save screenshots or PDFs with the test’s browser, viewport, URL, and build metadata. Compare like-for-like captures; a visual difference is a signal for investigation, not automatically a defect.
Or skip the browser setup
ScreenshotNeo provides a website screenshot API and MCP server. A single request can capture PNG, JPEG, WebP, or PDF output:
Quick Recap
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpBefore capture it accepts cookie or consent banners and removes more than 60 known consent platforms, newsletter popups, and chat widgets; each step can be turned off. Bot checks or CAPTCHAs, blank pages, timeouts, failed loads, and cache hits are not billed, and response headers report the page verdict and billing status. Its MCP tools—take_screenshot, get_page_info, and capture_pdf—let Claude, Cursor, or another MCP client capture pages. The Free plan includes 1,000 shots per month with no card; paid plans start at $5 for 3,000 shots. Create a free ScreenshotNeo account.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
