2
8 Comments

Has anyone run into trouble related to scalability with Bubble?

Hi! Was looking for an interviews with founders that used bubble to build their tools, but didn't find anything useful.

So I'm wondering, can anyone here speak from first hand experience about how well bubble scales? Is it just good to use for building an MVP or can you get to 100,000s of users before running into problems?

  1. 1

    I've recently been involved in an app build that scaled to >10k users. Bubble can scale with you, it's just a matter of knowing how to build for the outcome you're anticipating. The best answer unfortunately is "it depends"... and it can depend on a lot as every app is different.

    (stands on soap box)
    Many people build quickly on Bubble (without pausing to consider software design best practices) to rapidly get to "launch" to answer the question "do people want this thing". That approach totally makes sense -- why spend 6 weeks building something if you can build a crappy (but functional) first version in a weekend. If you take that shortcut though, you're likely taking on "technical debt" (ie complexity) that you have to pay off in the future, otherwise your users will pay the price with slow loading pages. If you want to build your app long term, this is in your favor as well as it's painful to maintain a "visual programming" app without automated tests and thorough documentation.

    Some tips:

    • If you're experiencing slowness in your Bubble app, open the debug tool (add ?debug_mode=true to the URL) and step through each of your actions on page load and with key interactions.
    • If your page or workflows are comprised of long workflows with complex "do a search for" lookups or complex Repeating Groups, that's likely a source of "slowness"

    Some quick wins for performance:

    • in Repeating Groups consider using "ext. vertical scrolling" instead of "full list"
    • watch out for nested repeating groups (ie a RG within a RG)
    • buy more "units" of capacity (comes at a cost though)

    More complex approaches

    • consider using hidden groups and the Router plugin to simulate page navigation instead of doing separate pages. This is best practice for mobile-first web apps.
    • create new data types that are tailor made for complex queries (ie aggregations)
    • store referenced data type on data types to avoid (ie store "Likes" on a "User" profile instead of having to search for Likes by a User)

    And give these a read
    https://manual.bubble.io/architecture-optimization-and-limits-of-the-bubble-engine/performance-and-scaling
    https://forum.bubble.io/t/a-fair-and-honest-chat-about-performance/35685/3?u=eric3

    If you're looking to build an app that's high performance and high polish, check out some we've built at Imua for examples of what's possible.

    https://imua.studio/work/uneebo?utm=indiehackers

    Happy building :)

    1. 1

      awesome! thanks for the detailed advice!

  2. 1

    So I just heard the podcast. It was actually their this week in startups episode 1021 starting at 18:40. He said the Twitter clone someone made on bubble can scale to hundreds of thousands of users.

    1. 2

      Thanks. I had listened to the first half of another interview on Indie Worldwide. Really appreciate you following up with this question!

      1. 1

        No problem! Sorry for sending you to the wrong place!

  3. 1

    I’m also keen to hear more on this, it’d be great to hear from those that have walked this path. From their forums, generally the issues with scalability have come down to poor database or workflow structuring.

    As a specific example, when you do a search in a repeating group using a constraint in the search rather than running a filter, can make a big difference.

    I also read the other day they’ve just launched a performance improvement to do with how the backend manages workflows which should help a load.

  4. 1

    I believe in their podcast interview with indie worldwide they say you can build an app to 100,000 users. You might have to listen to it and double check though.

    1. 1

      thanks, will check it out.

Trending on Indie Hackers
Aim to be valuable and you'll be indispensable. 24 comments I made $804 in February 17 comments How hard should you work? 16 comments NFTs are a dangerous trap 11 comments Tesla closes its forums and raises the anger of fans 6 comments How early-stage startups can build a strong Customer Feedback Loop 3 comments