Why is my website slow?

Rarely because of one big fault. More often it is many small ones: a server without caching, photos straight from the phone, and plugins that each load their own files. Our test finds the signs that can be read in the front page’s HTML.

Last checked:

What we measure

Our server fetches the front page’s HTML, the text and the recipe for the page, but not images, scripts or fonts. From that one file we look at six things.

Response time

How many milliseconds the server took to deliver the HTML, measured from our server. Over 1.5 seconds we call slow.

HTML size

More than about 300 KB of uncompressed HTML usually comes from page builders or data written straight into the page.

Compression

Whether the HTML is sent packed with gzip or Brotli. It typically cuts text files to a third of their size or less.

Files and other servers

The number of script files, inline scripts, stylesheets and images, and how many other servers the page loads from.

Render-blocking scripts

Scripts in the page head without async or defer. The browser waits for them before it shows anything. More than three gives a warning.

Images without size

Images without width and height make the content jump while the page finishes.

What we do not measure

We do not start a browser. So we do not know how long the whole page takes to finish, how many MB the images weigh, or how it does on Core Web Vitals (LCP, INP and CLS). For that, use Google PageSpeed Insights. Use our test to find the technical causes and PageSpeed Insights to measure the loading itself.

Test the speed of your website

Free. We read the front page and the domain's public DNS, the way a browser and a mail server would. Nothing is changed on the site and nothing is e-mailed.

When the server is slow to answer

If the server spends more than a second on the HTML, it is usually building the whole page from scratch on every visit. On WordPress that means PHP and the database work for every single visitor. A page cache stores the finished page and sends it straight away, and on most small WordPress sites it is the biggest single improvement. Many hosts have caching built in; otherwise a caching plugin can do it.

If caching does not help, look at the host. Cheap shared servers share their power with many other sites. Measure several times at different times of day before you decide to move.

Typical causes on small WordPress and Shopify sites

  1. Large images. A phone photo often weighs several MB. Make it no wider than it is shown, and save it as WebP or a compressed JPEG.
  2. Too many plugins or apps. Every WordPress plugin and every Shopify app can add its own scripts and stylesheets to every page. We flag more than 40 script files. Remove what you do not use.
  3. Scripts that block. Add defer where the theme or plugin allows it. Many caching plugins can do this, but click through the site afterwards, because a script can stop working.
  4. Other servers. Chat windows, tracking scripts, fonts and embedded videos each come from their own server, and every new server costs an extra connection. Many also set cookies, so they belong behind the consent banner.
  5. Heavy page builders. Page builders wrap every block in layers of code, so the HTML grows. A lighter theme or fewer sections on the front page helps.
  6. No compression. gzip or Brotli is switched on in the host’s control panel or on the server.

On Shopify, Shopify runs the server, so there it is mostly about apps, the theme and images.

The number is ours, not your customer’s. The response time is measured once from our server, over our connection. A customer on a phone somewhere else gets a different number, and one slow measurement can be chance. We test the front page only.

Questions about website speed

Why do I get the same number when I test again?

The same address within fifteen minutes is answered from our memory, so the site does not get needless visits. If you have just switched on caching, wait 15 minutes and test again.

What is a good response time?

We flag anything over 1.5 seconds. A page served from cache is typically far below that. More important than one number is whether it stays low over several measurements.

Do more plugins always make the site slower?

No. It depends on what they do. One plugin that loads five scripts from other servers can cost more than ten small ones. Look at the number of files and servers in the report.

Why is the PageSpeed number different from yours?

Because we measure something else. We only look at the HTML and the server’s answer; PageSpeed Insights loads the whole page in a browser.

Should I change host?

Only when caching, images and plugins are sorted and the response time is still high over several measurements. Otherwise you just move the problem with you.

Fixed price, no commitment

Get it fixed

Tell us which website it is and what the check showed. A person answers with a fixed price before any work starts.

  • A fixed price before we start
  • An answer by e-mail, no calls
  • Your details are used only to answer you

Fields marked with * are required.

What should be fixed?