Alright speed fans, listen up. I’ve not got the solution to our Google Pagespeed problem, but I’ve been doing some research on my app and I’ve found some major points of improvement.
One of the most annoying things is I’m being told by the debug mode that my app is being slowed down by 2MB of searches, but you can’t find out where they’re coming from. I really don’t have a lot of searches in my app, and they’re hidden away somewhere that I can’t remember.
Now this process might seem rudimentary to some of my more developer-minded colleagues, but I’ve got no real coding or HTML web dev experience so this is how I rooted the big searches out:
Load your app with your developer tools switched on (I use Chrome, I find it easier). In the Network tab, look at the log of things being loaded. If you organise by descending size, you might see a few ‘mget’ lines. I had a few, for example, at like 700KB, 400KB, 300KB, etc. Click on the mget line to see more details, then go to the “Preview” tab, you’ll see a list of all the things being returned. Expand them to get to the unique ID of what’s being returned, then find out what that corresponds to in your app.
I discovered there’s a lot of data being returned when for example I set a condition like "When Current User’s School’s (1 item) Groups’ (10 items) Lessons (100 items) Posts:sorted(by creation date):first item (1 item)… ". I guess it needs to load up all results at each level until it gets to the Thing you’re looking for (can anyone with more Bubble knowledge confirm this??). When I set another condition to “When This Lesson’s Posts:sorted:first item…” it barely loads any data at all, comparatively.
Anyway, I hope this has made sense. I’m going on a major hunt for data hungry conditions and filters now. Basically, if you have a condition that requires calling a lot of levels of data to find one Thing, it might be sucking up your speed.
I think the issue is the slow speeds occurring even when there are no conditions or filters. If you see my post above an empty page on mobile scored 14 /100. So our request to bubble is to look at infrastructure or design. and not specific to any one application.
Yeah for sure, sorry I didn’t mean to say that fixing the conditions / filters would help the Google score, just that it sped up the actual page load for my users.
The problem with Bubble is like having a Wordpress installation with no speed-boosting plugins like W3 Total Cache to control render-blocking scripts etc. I’m not sure that’ll be fixed anytime soon.
I am still working to gain performance. Has anyone used CloudFare and thinks that it helps with bubble page load speeds ?
I tried CloudFare but the site went down a couple of times and couldn’t establish if its an issue on CloudFare or Bubble end. So ended up removing CloudFare. Any success stories or product recommendations to optimise speed would be appreciated.
Thanks for the reply and affirming the continued efforts in this performance sector.
Interesting with regards to a big player such as Airbnb, they seem to be hovering in the low 30-50 range, I’ve found that the results do vary.
The site I’m working on is a 5, granted there are a few things I can improve here as its the homepage with fair amount of content. Anyhow, hope to see the continual improvements overtime - please keep us posted as to any updates that has made an impact. Thanks again.
