Short answer: a CDN is worth keeping only if the numbers for your site improve for your visitors. You do not decide it by hearsay: you measure before, switch it on, and measure after with the same yardstick. If there is no gain, switch it off and you have lost nothing.
We do not sell a CDN. The one a customer can use is Cloudflare, on an account of your own, opened directly with them. What a CDN is and when it helps is in what a CDN is and when it makes a difference. This article answers the next question: how do I know whether it really made a difference in my case?
What changes, side by side
|
Without a CDN |
With a CDN (Cloudflare) |
| Images, CSS and JavaScript |
All served from your server, to every visitor. |
After the first request, served from a copy on a machine near the visitor. |
| The page itself (WordPress, PHP) |
Built on your server. |
By default still built on your server. According to Cloudflare’s documentation, only static files are cached by default, not the HTML. |
| Visitor far from the server |
Every file makes the whole trip. |
Static files come from nearby. This is where the gain shows. |
| Visitor near the server |
The trip was already short. |
A small gain, sometimes indistinguishable from noise. |
| If something goes wrong |
There is one place to look. |
There is an extra layer, and the cache may show an old version. |
The test, in six steps
| 1 |
Pick three pages and do not change your mind halfway: the home page, one with many images and one inner page.
|
|
| 2 |
Measure without the CDN. Always use the same tool and, if it allows, a test location close to where your visitors are. Repeat each page at least three times and keep the middle value. Note the time to first byte and the time until the page is complete. The tools and what the numbers mean are in measuring your site speed.
|
|
| 4 |
Let the cache warm up. The first request for each file goes to your server; only from the second one does it come from the copy. Visit the same three pages a few times before measuring.
|
|
| 5 |
Confirm the cache is working. In a terminal, curl -I followed by the address of an image on your site shows the response headers. On Cloudflare, the cf-cache-status header says HIT when the copy came from the CDN and MISS when it had to fetch the file from your server (according to Cloudflare’s documentation). If you never see HIT, there is nothing to compare.
|
|
| 6 |
Measure again, exactly as in step 2, and put the two sets of numbers side by side.
|
|
|
The commonest mistake: measuring once in the morning without the CDN and once in the afternoon with it. The normal variation between two runs can be larger than the gain you are looking for. And on a WordPress site without a page cache, the time to first byte barely moves: a CDN shortens the journey, it does not speed up building the page.
|
How to read the result
| What you see |
What it means |
What to do |
| Static files faster, first byte the same |
The normal result. The CDN did what it says. |
Keep it, especially if your visitors are far away. |
| Nothing changed |
Your visitors are close, or the site is already light. The CDN was not the bottleneck. |
Keep it or switch it off. Look for the cause elsewhere: what actually makes a difference. |
| It got worse |
A badly chosen cache rule or SSL mode, or files that never reach HIT. |
Review SSL at Cloudflare and, if that does not solve it, switch the CDN off and confirm it is off. |
|
If the site is only slow for you and not for others, the problem may not be the server at all: see fast for you, slow for everyone else. And if what worries you is protection from unwanted traffic rather than speed, the test is a different one: what the CDN lets through and what it blocks.
|
|
Not sure where to start measuring? Tell us which page is slow and what you have already tried.
Open a support ticket
|
RECOMMENDED PRODUCT Web hosting with cPanel Domain and SSL included, daily backups and the panel you already know. from $6.59/mo (3-year plan, with coupon) See plans |
Português
English
