One of my websites has several articles with embedded YouTube videos.
I wanted to find out if embedding a video slows down the website. So, I ran a series of before-and-after tests to measure the impact on one of my websites.
Here is my verdict.
Table of Contents
Does Embedding Video Slow Down Your Website?
When you embed a video on your website from a hosted platform like YouTube or Wistia, visitors to the website should not experience a noticeable slowness.
Performance issues are most likely to occur if the website is on entry-level hosting with low resources.
The most common method of embedding videos is to use YouTube as the platform. In a series of tests run for this article, we found that adding a YouTube video to a web page produced a negligible decrease in perceived speed.
There’s no doubt that a significant number of extra scripts and calls are added to the website by the embedded video. However, these scripts are run asynchronously.
In other words, your visitors can interact with the site while scripts are loading in the background. This is how you can have embedded videos without any perceived slowness.
Our Embedded Video Performance Tests
I tested performance using the GTMetrix tool on the same 800-word web page under three conditions:
- No images, no embedded video
- No images, one embedded video from YouTube
- Six images, one embedded video from YouTube
Here are the results:
Images | Video | Time To Interactive (ms) | Fully Loaded Time (ms) |
None | None | 346 | 581 |
None | 1 | 407 | 3,200 |
6 | 1 | 933 | 3,800 |
What do “Time To Interactive” And “Fully Loaded Time” mean?
Here’s a quick explanation of the measurements which are in milliseconds.
“Time To Interactive” basically means how long it takes for the web visitor to be able to scroll around the page.
It’s more of a “quality” metric. If it’s above two seconds, we should be worried that the user will perceive that the website is slow.
“Fully Loaded Time” is when all the resources have been loaded by the page. That includes every image, all scripts, and any external calls to other sites.
You should focus on the “Time To Interactive” metric instead of the “Fully Loaded Time”.
It hardly matters if there’s a script still loading while the visitor is absorbing the contents of the first half of the page. As long as this metric is below 2000 ms, then your page will not be perceived as slow.
With that explained, let’s take a closer look at the numbers from this test.
Test results
Don’t be alarmed by the significant jump in the “Fully Loaded Time”. Surely a figure of three seconds is a problem? Well, not if the visitor doesn’t notice that scripts are loading in the background.
Focus instead of the “Time To Interactive” metric. As you can see, simply embedding the video had a negligible impact on this measure.
On the other hand, adding a further six images slowed the page a little. The “Time To Interactive” measure more than doubled.
However, it was still below a full second. This will not be perceived as a slow page.
How Does Embedding A Video Impact Your Website?
Despite not having to worry too much about the increase in the Fully Loaded Time, you should still understand why it increases significantly when you embed a video from YouTube.
Here is a side-by-side comparison of the resources loaded on a web page that doesn’t have an embedded video versus one that does.
Where I’ve blanked out a domain name, that’s simply the website itself. I’m showing all the external websites involved.
You may need to expand this to see it clearly, but I’ll explain the main points.
The page without a video makes 12 requests. Adding a video results in an extra 22 requests. I’m not actually showing all the requests on the right side.
But you can see that many are to YouTube.com and some are to other Google sites.
I don’t have advertising on the website I used for testing. However, embedding YouTube means that extra scripts for Google’s advertising get called.
How to Check If Embedded Videos Are Slowing Your Website
If you have recently embedded videos and are seeing a marked slowdown on your website, then the first step is to check that the videos are actually the culprit.
Something else may have hit the performance.
I use the free GTMetrix site to check performance. There are alternatives, including Google’s lighthouse feature (GTMetrix also uses this).
You can take these steps to assess the impact:
- Run the page with embedded video through the speed test a few times (get a few measures)
- Remove the video and publish (this is temporary)
- Run the page without embedded video through the speed test a few times
If you don’t see any massive difference, then it’s probably not the video. Go put it back!
What To Do If Embedded Videos Are Slowing Your Website
If your site is on cheap entry-level hosting, then you may have been allocated very sparse resources by your hosting provider.
One potential solution is to upgrade your hosting to a more powerful package.
If that’s not an option, then you should take a closer look at how you are embedding the video.
Alternatives to YouTube
YouTube has the great advantage of being free to use and easy to set up. But as we showed in a previous section, embedding a YouTube video will pull extra advertising scripts into your site.
And unless you own the YouTube channel, you’re getting no monetary benefit from those scripts.
Wistia is an alternative video hosting service that offers a more lightweight option for embedding your videos.
Wistia isn’t free, but you may find that the costs are worth the faster loading speeds.