Free tools Windows power users keep installed
One-click scans. No signup required.
To properly test a website’s speed, start with the exact page you want to evaluate in Google PageSpeed Insights (PSI). Read its real-user field data separately from its simulated Lighthouse lab results, then use Lighthouse, Chrome DevTools, or WebPageTest to investigate specific bottlenecks. A single score is not a complete measure of how visitors experience a site.
Run a repeatable website speed test
- Choose the page and audience. Test the landing page, product page, or other URL that matters—not just the home page. PSI can report data for an individual URL; when it lacks enough samples, PSI may use origin-level data instead. Google explains how PSI reports field and lab data.
- Run the URL through PageSpeed Insights. Record the mobile or desktop view you are examining, the Core Web Vitals and percentile shown, and whether the field report is URL-level or origin-level.
- Keep field and lab results separate. PSI’s field data comes from the Chrome User Experience Report (CrUX), when available, and reflects real-user experience over a trailing 28-day period. Its lab data comes from a simulated Lighthouse page load. They answer different questions and are not expected to match exactly. PSI documentation.
- Choose a diagnostic tool for the problem. Use Lighthouse or Chrome DevTools for lab audits and local runtime debugging. Use WebPageTest for controlled comparisons and deeper performance inspection. These tools complement PSI; their readings are not interchangeable. Google’s Mobile Site Speed Playbook.
- Repeat comparable lab tests before judging a change. Keep relevant conditions as similar as practical. Network availability, client hardware availability, and client resource contention can cause run-to-run variation, so do not attribute a small score shift to a code change based on one run. Google’s PSI documentation.
- Check field data after deployment. Revisit PSI for the longer-term real-user view. Because CrUX uses a trailing 28-day window, field data will not behave like an immediate before-and-after lab test. The Lighthouse project describes how to use lab and field data.
Understand what the results mean
Core Web Vitals are the user-experience measures to inspect
Google’s 2024 guidance classifies field experience at the 75th percentile. The good thresholds are Largest Contentful Paint (LCP) of 2.5 seconds or less, Interaction to Next Paint (INP) of 200 milliseconds or less, and Cumulative Layout Shift (CLS) of 0.1 or less. An aggregation with sufficient data for all three passes when all three are rated good. These are metric thresholds, not a claim about how many websites meet them. Google’s Web Vitals guidance.
As an Amazon Associate I earn from qualifying purchases.
A high Lighthouse score is not the same as passing real-user data
PSI labels a Lighthouse Performance category score of 90 or higher as good. That is a lab result: Google cautions that a good lab result does not guarantee a good real-user experience. Do not use a 90-plus score as a substitute for checking field Core Web Vitals. Google’s PSI documentation.
Missing field data is inconclusive
If PSI does not show page-level field data, the URL may be new or may not have enough real-user samples. PSI may then show origin-level data; if the origin also lacks sufficient samples, it cannot report real-user experience data. A missing field report does not establish that the page is fast or slow. Google’s PSI documentation.
#1 Best Overall
Choose the right speed-testing tool
| Tool | Best use | What it tells you |
|---|---|---|
| PageSpeed Insights | Quick check of a public URL | Combines Lighthouse lab diagnostics with CrUX field data when enough real-user samples are available. Google documentation. |
| Lighthouse | Immediate lab audits and debugging | Simulates a page load and provides lab feedback; it does not replace real-user field data. Google’s Mobile Site Speed Playbook. |
| Chrome DevTools | Investigating runtime behavior locally | Its performance tools help profile execution and investigate bottlenecks. Google’s Mobile Site Speed Playbook. |
| WebPageTest | Controlled comparisons and deeper investigation | Supports deeper performance inspection and can also run Lighthouse. Google’s Mobile Site Speed Playbook. |
| Search Console Core Web Vitals report | Site-level view of real-world usage | Provides a site-level perspective, unlike PSI’s individual-URL check. For an individual URL, start with PSI. |
How to compare a speed improvement fairly
- Use the same URL and device category before and after the change.
- Compare like with like: lab runs against lab runs, and field data against field data.
- Repeat lab measurements when a difference is small; a single run can vary with available network, hardware, or client resources.
- For field results, allow for the trailing 28-day CrUX window rather than expecting an instant change after deployment.
Google summarizes the distinction this way: “Lab data offers reproducible results and a debugging environment, but might not capture real-world bottlenecks and cannot correlate against real-world page KPIs.” For lab iteration and long-term monitoring, the Lighthouse project recommends using lab data to improve the experience and field data to understand users’ experience over time. Google’s Mobile Site Speed Playbook; Lighthouse documentation.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How many websites pass speed benchmarks?
The cited Google guidance establishes thresholds and explains how to measure performance, but it does not establish a representative current percentage of websites that pass. Treat claims about a universal site pass rate cautiously unless they identify a suitable, current population and measurement method.
Quick Recap
Best Value
Rank #3
Rank #2
- Used Book in Good Condition
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.
