I recently pieced together a frontend web app which uses supabase for the backend. However I realised that others had posted similar topics to this - but used the “public” schema, instead of a custom schema.
I’ve attached some images of the settings I used for my API Connector in case that helps.
also I have attached links to my in-depth guide if you prefer a video presentation.
I apologise for the lack of context in these photos - if you require more context please visit my youtube/blog post
What advantages does this have over just using native Bubble DB? I can understand using Supabase if you are using the plugin which saves on WUs but unclear on how using the API Connector would advantage a bubble dev over just using native Bubble?
To answer your question, there isn’t an overwhelming advantage to using the native Bubble database over Supabase.
However, one appropriate use case would be if your backend logic has grown too complex for Bubble’s native capabilities.
If you want granular control over your data and are familiar with SQL commands, you can create some impressive backend functionalities, such as functions that return complex calculations, which might be more challenging to achieve with Bubble’s frontend alone.
It’s also a personal preference; I feel more comfortable and have greater insight into my data when using SQL commands.
to piggyback off of @GH5T I haven’t explored the performance aspect of what data Im pulling in vs webapp speed. For most small applications we shouldn’t need to worry about this aspect (yet!)
Personally, I believe any query outside of Bubble is cheaper and more performant (I have not concluded any tests myself), but with Supabase and the serverless capabilities however, I can almost assume a landslide competition. The reason I say this is because of how much they charge per search, or API request. Supabase charges a fraction of that – if not free.
Any other external db or backend more way faster and perfomant ( but takes time to integrate )
Just yesterday i had to transfer my workflows ( complex calculation per row , summing totals, displaying result to users) to AWS Lambda as it was too slow on Bubble. Even with Schedule API List.
I think there is only one good reason to keep using bubble db. Real time updates. Supabase can do this as well, not very hard but you need to put in a bit of time. Half a day or so.
Other than that I think Supabase only has advantages and more now considering how good AI is when asking to write complex functions.
Also cheaper
Also performs better
Also the ability to use complex functions
Also easy to use logical replication out of the box
Also easy to find devs as it is “standard” Postgres
Also no lockin
Also you remain owner of the data
Also more flexibility as you can easily swap one frontend for the other
Hey thanks for the contribution
Can you elaborate on what you mean by real time updates? If a user updates a value isnt that real time sorry I may have mistunderstood you
Real-time updates in “Bubble Terms” is a websocket that can read new data at any given time without page reload.
Bubble does this out of the box, but when you add in your own DB, that process is killed. You must implement your own websocket or refresh capabilities. It’s really not that hard to implement. There are some guides on here that solve that problem.
