New! Free AI Visibility Report: does ChatGPT recommend you? →We work US · UK · CA · AU hours
Web Design, CRO and UX

How to Test and Measure Website Performance and Speed

QBy QHR Solutions team··3 min read
Laptop showing website performance analytics representing website speed testing and optimization

Website performance affects how quickly visitors can view content, interact with forms, and complete important tasks. Testing speed regularly helps you identify bottlenecks and check whether changes to images, scripts, hosting, or page design are improving the experience.

Start With the Right Testing Tools

  • PageSpeed Insights: Provides lab diagnostics and, when available, real-user field data for a page.
  • Google Search Console: The Core Web Vitals report can help identify groups of URLs with user-experience issues when enough field data is available.
  • Web Vitals guidance: Explains the metrics and recommended thresholds.
  • Browser developer tools: Network and performance panels help developers investigate large resources, JavaScript work, and rendering delays.

Understand the Core Web Vitals

Google’s Core Web Vitals focus on three aspects of real user experience:

  • Largest Contentful Paint (LCP): How quickly the main visible content loads. A good result is 2.5 seconds or less.
  • Interaction to Next Paint (INP): How responsive the page is to user interactions. A good result is 200 milliseconds or less.
  • Cumulative Layout Shift (CLS): How much unexpected layout movement occurs. A good result is 0.1 or less.

These thresholds are assessed at the 75th percentile of page visits, segmented across mobile and desktop. Lab tests are useful for diagnosis, but field data reflects the conditions experienced by real visitors and may not be available for every URL.

Test Representative Pages

Do not test only the homepage. Include key service pages, blog posts, product or category pages, and important conversion pages. Test mobile and desktop separately, and repeat tests under reasonably consistent conditions before comparing results. Third-party scripts, caching, network conditions, and test location can affect scores.

Find the Main Bottlenecks

Large images and media

Resize images to the dimensions they are displayed at, use modern formats such as WebP or AVIF where supported, and compress files without visibly damaging quality. Avoid lazy-loading the main above-the-fold image if it is the LCP element; lazy-load off-screen images instead.

JavaScript and third-party tools

Review analytics tags, chat widgets, advertising scripts, and unused code. Remove scripts only after checking whether the business relies on them. Defer or delay non-critical work when appropriate and test forms, checkout, and tracking after each change.

Fonts, CSS, and layout shifts

Reduce unnecessary font variants and CSS, reserve space for images and embeds, and avoid inserting banners above existing content without allocating room. These changes can improve loading and reduce unexpected movement.

Server response and caching

Check hosting response time, caching configuration, database queries, and content delivery for your audience. A cache or performance plugin should be configured for the specific platform; conflicting optimisation features can break layouts or functionality.

Prioritise Fixes by Impact

  1. Record the current results for key templates and devices.
  2. Identify the largest bottleneck and its likely cause.
  3. Make one controlled change where possible.
  4. Retest the same pages and verify that functionality still works.
  5. Review field data over time to see whether real users benefit.

Do not optimise a score in isolation. A change that improves a laboratory score but breaks navigation, accessibility, analytics, or checkout is not an improvement. Page speed is one part of technical SEO and user experience, not a guarantee of higher rankings by itself.

Frequently Asked Questions

Why do PageSpeed scores change?

Test conditions, caching, third-party scripts, server load, and the resources loaded by the page can vary. Compare repeated tests and use field data where available.

Should every page reach a score of 100?

No. Aim to resolve meaningful performance issues while preserving accessibility, reliability, and essential features. A perfect score is not required to provide a good user experience.

How often should a website be tested?

Test after major design, theme, plugin, script, or hosting changes, and review important pages periodically. Ecommerce and lead-generation pages deserve particular attention because performance problems can interrupt key actions.

Keep Performance Under Review

Use a repeatable process: measure, diagnose, implement, and validate. For broader technical improvements, see our page speed optimisation guide and website maintenance resources.

Q
Written by the QHR Solutions team

Our SEO, ads and web specialists share what works for clients in the US, UK, Canada, Australia and UAE. About us →

Liked this? You'll love results.

Let's grow your business.

Free 30-minute strategy call with a senior strategist. No contracts, no pressure.

Book My Free Call →