Website Speed Test Tools: The Complete Guide

$body$

"My site is slow" is not a diagnosis — it is a symptom. To know what to fix, you first have to measure. The good news: the best speed testing tools are free. The less good news: most people use them wrong and chase the wrong numbers.

PageSpeed Insights — first and mandatory

Google's tool (pagespeed.web.dev) gives you two kinds of data. Lab data (the Lighthouse score, 0–100) — a simulation on a standardized device. And field data (CrUX) — real experiences of your actual visitors over the last 28 days. The latter matters more: Google uses field Core Web Vitals for ranking, not the lab score.

The key numbers that should be "green": LCP under 2.5 seconds, INP under 200 milliseconds, CLS under 0.1. If your site does not have enough traffic for CrUX data, rely on lab results — but with a grain of salt.

GTmetrix — detailed waterfall analysis

GTmetrix (free with registration) shows what PageSpeed does not present clearly enough: the waterfall chart — a timeline of every single file loading. There you can see exactly what is holding things up: a slow server response, a huge image, a blocking third-party script. Tip: set the test location closest to your audience — the default Canadian location gives unrealistic numbers for European visitors.

WebPageTest — for serious diagnostics

WebPageTest.org is the most powerful of the free tools: real devices, network selection (3G, 4G, cable), before/after comparisons, and a filmstrip view showing frame by frame what the visitor sees. When you want to prove whether a change actually helped — this is the tool.

Other useful tools

• Chrome DevTools (Lighthouse and the Performance tab): local testing with simulated slower networks and CPUs.
• KeyCDN / Pingdom Tools: quick tests from multiple global locations — useful for checking whether your CDN is doing its job.
• Search Console — Core Web Vitals report: shows which groups of pages fail the thresholds, based on real users.

The most common testing mistakes

• One measurement = a conclusion. Results vary; test at least 3 times and look at the median.
• Testing only the homepage. Visitors also land on blog posts, products, contact pages — test those too.
• Chasing a score of 100. The difference between 92 and 100 is invisible to visitors; the difference between 40 and 85 is not. The score is a guide, not a goal.
• Testing a cached version only. The first test "warms up" the cache; the second shows reality for most visitors. Look at both.
• Ignoring TTFB. If time to first byte is over 600 ms, no front-end optimization will save the result — the problem is hosting or caching.

When the numbers point at the server

High TTFB, slow first response, good results only from cache: those are signs the bottleneck is your hosting. NVMe storage, server-side caching, and a CDN are included in all our plans, and for WordPress sites we offer preconfigured WordPress hosting. Measure before, migrate, measure after — the numbers will speak for themselves.

Finally, a practical rhythm: PageSpeed Insights monthly as a checkup, GTmetrix whenever you change something, WebPageTest when hunting for a cause. Measure, change one thing, measure again — that is the whole recipe.

$body$