Free tool ยท No sign-up

Website Performance Checker

Enter any page and get a scored performance audit: how fast the server answers, how heavy the page is, which files block rendering, whether images and scripts are compressed and cached, and what to fix first.

  • Scored report with a fix list
  • Measures real files on the server
  • Server, weight, images and caching

For real-user speed data, use the Core Web Vitals Checker. To go further on pictures, try the Image SEO Checker.

What it checks

Five areas that decide how fast a page feels

Each area gets its own score, so you can see where the time and the bytes are going.

Server and delivery

How long the server takes to start answering, whether the HTML is compressed, how many redirects come first, and whether the page is served over https.

Render blocking

Scripts and stylesheets in the head that stop the page from painting, CSS @import chains, and web fonts that hide text while they load.

Page weight

How many files the page pulls in, how many bytes they add up to, how big the DOM is, and how many third-party sites are involved.

Images

Pictures with no width and height, a lazy-loaded hero image, images that are far too large, and JPEGs and PNGs that should be WebP or AVIF.

Caching and compression

Whether CSS, JavaScript, fonts and images are compressed and cached for a long time, so returning visitors don't download them again.

Overall score

The five areas are combined into one score and a grade from A to F, with a prioritised list of what to fix.

Targets

Performance targets to aim for

These are widely used guidelines. The tool rates your page against them.

Website performance targets
MeasureGoodNeeds workWhy it matters
Server response (TTFB)Under 200 msOver 800 msNothing can paint until the server answers.
Total page weightUnder 1 MBOver 3 MBHeavy pages are slow on mobile data.
JavaScriptUnder 300 KBOver 1 MBScripts must be parsed and run, not only downloaded.
Single imageUnder 200 KBOver 500 KBImages are usually the largest files.
DOM elementsUnder 1,500Over 3,000Big pages slow rendering and every interaction.
Render-blocking files0 scripts3 or moreEach one delays the first paint.
Cache lifetime for assets30 days or moreNoneRepeat visits should not re-download files.
How to use it

How to check website performance

Three steps from a URL to a fix list.

1

Enter a URL

Paste any public page. Leave file measuring on to include sizes, compression and caching.

2

Read the score

Start with the lowest area score, then the errors in the list. Each finding says how to fix it.

3

Fix and test again

Work from the errors down, deploy, and run the check again to see the score rise.

Quick wins

The fixes that usually matter most

On most sites, a handful of changes make most of the difference.

Compress and resize images

Resize pictures to the size shown, save them as WebP or AVIF, and lazy-load those below the first screen.

Defer scripts

Add defer to head scripts and load analytics and chat widgets after the page is usable.

Use a CDN and caching

Serve the page and its files from a CDN with long cache lifetimes on versioned files.

Turn on Brotli or gzip

Compressing text files typically cuts them by 70 percent or more.

Remove what you don't use

Delete unused plugins, fonts, libraries and tracking tags. The fastest file is the one you don't send.

Load fonts well

Add font-display: swap, preload the main font and limit the number of weights.

Why it matters

What a website performance checker does and why page speed matters

Website performance is how quickly a page is delivered and shown, and how much it costs the visitor to load. It starts on the server, where the first byte of the response has to be produced and sent. It continues in the browser, which downloads the HTML, then the stylesheets, scripts, fonts and images the HTML points to. Anything that makes that chain longer or heavier slows the page.

A website performance checker looks at that chain and points to the links that are weak. This one does not run the page in a browser. It reads the server's response and the files the HTML references, so it can name exact causes: a slow first response, HTML sent uncompressed, scripts in the head that block the HTML parser, images far larger than they are shown, and static files with no cache lifetime.

Speed matters for people first: slow pages lose visitors and enquiries. For search, Google's measurements of real-user experience are the Core Web Vitals, and this tool is the diagnostic step behind them. A slow server response, for example, is not a Core Web Vital, but it takes time away from your Largest Contentful Paint. Use the Core Web Vitals Checker to see what real visitors experience and this tool to see what is causing it.

When to use it

When to run a website performance check

Run it whenever the server, the theme or the files a page loads have changed.

Before a site goes live

Test the staging or pre-launch URL for uncompressed files, missing cache headers, large images and blocking scripts. These are cheap to fix before launch and awkward to fix later.

When comparing hosting or a CDN

Check the same page before and after a move. The server response time is measured from Cloudflare's network, so use it to compare setups, not as a stopwatch for your own visitors.

After adding a plugin or tag

A new slider, chat widget or tracking tag adds files and requests. Run the check again and see whether weight, third-party hosts or blocking files went up.

When LCP is slow in the field

A poor Largest Contentful Paint often comes from a slow server, a blocking stylesheet or a heavy hero image. This report shows which of the three is the bigger problem.

During an image clean-up

See which images are far larger than needed, which lack width and height, and which are still JPEG or PNG when a modern format would be smaller.

In a technical audit or competitor study

Compare your page with a competitor's on weight, files and caching. The fix list gives a developer a place to start without opening the browser tools.

Common mistakes

Page speed mistakes and how to fix them

The same few causes slow most sites, and most are configuration problems rather than code problems.

No page caching on a CMS

What goes wrong: Every visit makes the CMS (WordPress, for example, runs PHP and database queries) build the same page again, so the server takes a long time to send the first byte, which then delays everything else.

The fix: Turn on full-page caching at the host, in a caching plugin or at the CDN. On WordPress, page caching is a common first fix for a slow server response.

Scripts that block the first paint

What goes wrong: Scripts in the head without defer or async stop the browser from reading further until each one has downloaded and run. Stylesheets also block rendering, so a large stylesheet delays the first paint too.

The fix: Add defer to scripts that need the page parsed first, or async to independent ones. Keep the CSS needed for the first screen small, and load the rest later.

Short cache lifetimes on static files

What goes wrong: CSS, JavaScript, fonts and images served with no cache lifetime are downloaded again on every visit, which makes returning visitors wait for files they already have.

The fix: Give versioned files a long Cache-Control: max-age, for example a year, and change the file name when the content changes. Keep HTML on no-cache so it stays fresh.

Compression missing or misapplied

What goes wrong: HTML, CSS and JavaScript sent as plain text are bigger than they need to be. The opposite mistake is compressing images that are already compressed, which wastes server time for no gain.

The fix: Enable Brotli or gzip for text files at the server or CDN, and leave JPEG, PNG, WebP and AVIF files alone. Check the Content-Encoding header to confirm it works.

Oversized and wrongly loaded images

What goes wrong: A 3000-pixel photo shown at 600 pixels wide, or a hero image set to lazy-load, makes the largest element on the page slow to appear.

The fix: Resize images to the largest size they are displayed at, use WebP or AVIF, give every image a width and height, and load only the images below the first screen lazily.

Too many fonts, plugins and third-party tags

What goes wrong: Each extra font weight, plugin asset and external script adds requests and weight, and third-party hosts add their own connection delays that you cannot speed up.

The fix: Remove what you do not use. Limit font families and weights, preload the main font, and load analytics, chat and ad tags after the page is usable.

Glossary

Website performance terms explained

TTFB (Time to First Byte)
The time from the request until the first byte of the response arrives. It covers server work and network travel. Google's web.dev guide treats 0.8 seconds or less as a rough good target.
Render-blocking resource
A stylesheet, or a script without defer or async in the head, that the browser must handle before it can paint the page. Deferring scripts and trimming critical CSS reduces the delay.
defer and async
Attributes on a script tag. Both download the file without pausing the parser. defer runs scripts in order after parsing, while async runs each one as soon as it arrives.
Page weight
The total bytes a page downloads, across HTML, CSS, JavaScript, fonts and images. Heavy pages cost more time and mobile data, and images are usually the largest share.
Content-Encoding
A response header naming the compression applied to the body, such as gzip or br (Brotli). Text files compress well. Images that are already compressed usually do not.
Cache-Control
A header that tells browsers and CDNs how long they may reuse a response. max-age sets freshness, no-cache forces revalidation before reuse, and no-store forbids storing it at all.
DOM size
The number of HTML elements on a page. A very large DOM takes longer to style and lay out, and it can make later interactions slower.
CDN (content delivery network)
A network of servers that keeps copies of your files close to visitors. It shortens travel distance, can cache pages and static files, and often handles compression too.
Madhav Kushwaha, SEO, AEO, GEO and digital marketing consultant in Kathmandu, Nepal

Built by

Madhav Kushwaha

SEO, AEO, GEO & Digital Marketing Consultant in Kathmandu, Nepal, and trainer at ODcrU Academy and TechAxis Institute. Most speed reports give a score and little else. I built this checker to show the causes behind it: the exact files that block rendering, the images that are too heavy, and the headers that are missing, so the fix list is something a developer can start on at once.

Website Performance Checker: Frequently Asked Questions

A website performance checker tests how quickly a page is delivered and rendered, and what slows it down. This one measures the server's response, the page's weight and structure, and the size, compression and caching of the files it loads, then scores the result and lists what to fix.
The page is scored from 0 to 100 in five areas: server and delivery, render blocking, page weight, images, and caching and compression. They are weighted and combined into one overall score and a grade from A to F. If file measuring is switched off, the caching area is left out and the others are rescaled.
No. Pages are fetched from Cloudflare's network, not from your device, so response times describe how the server answers a nearby data centre. Use them to compare pages and spot slow servers, not as a stopwatch for your own visitors.
Time to First Byte is how long the server takes to start sending the page. Under 200 ms is excellent, up to 800 ms is acceptable, and over 1.8 seconds is poor. Slow TTFB usually means no page caching, a slow database or distant hosting.
Browsers can't draw the page until scripts and stylesheets in the head have downloaded and run. Each blocking file delays the first paint. Adding defer or async to scripts, and keeping the CSS small, is one of the biggest speed wins.
The tool reads the Content-Encoding header the server sends. If a server or CDN compresses only for some clients, or hides the header, the tool can report it as missing. You can confirm with curl, and check that both gzip and Brotli are enabled on your host or CDN.
Up to 30 per run: render-blocking CSS and JavaScript first, then fonts, the main image, and the rest in page order. Pages with more files are noted, and the sample is chosen so the most important ones are always included.
No. Those tools run the page in a real browser and measure rendering. This checker does not render the page: it analyses the server's response and the files the HTML references, which is faster and shows exact causes. For Google's own measurements, use the Core Web Vitals Checker.
No. The page must be public. Sites that block automated requests with a firewall may refuse the check with a 403 or similar. If your page is private, run your browser's Lighthouse panel instead.
No. The page and its files are fetched to measure them, the analysis happens in your browser, and nothing is saved or shared.

More free SEO tools

Speed, status codes, markup and links all shape how search engines and visitors see a page. Check them one by one.

Call WhatsApp SEO Audit