• Resolved avz

    (@avz)


    Hi there,

    This plugin looks very promising. For us the performance is an issue.
    If we look to Google Pagespeed the plugin seems to cause a 20 point drop for us. From around 80 to around 60. Is this ‘issue’ known?

    We enabled ‘turbo modus’. Any other suggestions?
    Thanks!

Viewing 6 replies - 1 through 6 (of 6 total)
  • Plugin Author Hessel de Jong

    (@hesseldejong)

    Hi @avz,

    Can you share your URL with us? There can be large differences per setup, there might be some errors that are causing a worse score. I’d love to take a look, and I might be able to make some suggestions.

    Kind regards,
    Hessel

    Thread Starter avz

    (@avz)

    Hi Hessel,

    Thanks for the quick response. We are working on this site right now: https://shorturl.at/klrH6

    Plugin Author Hessel de Jong

    (@hesseldejong)

    Hi @avz,

    The scripts needed for Burst all load within 50ms on your website, and the scripts are loaded asynchronously. So they should not cause a significant decrease in page speed score.

    The most important part of improving your page speed score is by decreasing the server response. From my tests, ‘waiting for server response’ takes about 1400-1500ms.

    I’ll do some tests with an empty website with just Burst, and I will report back to you later this afternoon. However, I suspect that the observed differences may simply be due to variations in server performance.

    Kind regards,
    Hessel

    Plugin Author Hessel de Jong

    (@hesseldejong)

    Hi @avz,

    I conducted some tests and found that when ‘Cookieless’ is enabled, it actually decreases the page speed score.

    On the other hand, when ‘Cookieless’ is not enabled, the score improves. This is likely just a fluctuation in the server performance. However, this does give us valuable information about the location of the speed issue, which is when ‘Cookieless tracking’ is enabled.

    It’s important to note that tracking without cookies will always be slower than tracking with cookies. This is because cookieless tracking requires a fingerprinting library, which increases the file size and demands more resources to run the fingerprinting code. However, we will look into ways of improving the page speed for cookieless tracking. I can’t give you an exact date for when we have improved the script.

    For now, you could disable ‘Cookieless tracking’ and the performance issue should be fixed. In the Netherlands, there is no need for cookieless tracking as far as I understand. You can use tracking with cookies as long as the data is stored locally and anonymously, which Burst does. There is no need for a cookie banner, but you do need to mention it in your privacy statement.

    I hope this clears some things up.. We will look into the pagespeed issue with ‘Cookieless tracking’ enabled. Please let us know if you find other issues and we’d be happy to help.

    Kind regards,
    Hessel

    Below are my tests:

    Without Burst: https://pagespeed.web.dev/analysis/https-xious-oryx-cako-instawp-xyz/4zqwpt4qk7?form_factor=mobile

    With Burst: https://pagespeed.web.dev/analysis/https-xious-oryx-cako-instawp-xyz/ojtpx8g9z4?form_factor=mobile

    With Burst Cookieless: https://pagespeed.web.dev/analysis/https-xious-oryx-cako-instawp-xyz/17xxpre0lo?form_factor=mobile

    Thread Starter avz

    (@avz)

    Thanks! Really appreciate this.
    We disabled cookieless tracking, and indeed the score improves right away.

    Plugin Author Hessel de Jong

    (@hesseldejong)

    Great to hear @avz! Hope you will enjoy using Burst 🙂

Viewing 6 replies - 1 through 6 (of 6 total)
  • The topic ‘Performance’ is closed to new replies.