After my LinkedIn post about experimenting with no-code tools and seeing a steady stream of posts on how AI and no-code are transforming development, I decided it was time to try out a no-code approach myself. Bubble.io was my first choice, but I suspect many of my findings could apply to other no-code platforms too.

The journey to 'production-ready': spoiler alert — it’s complicated

For starters, let’s talk about Bubble.io's database limitations. On the surface, things seem fine. You can have up to 10,000 rows in an element with a 1

relation. That sounds generous until you realize that, with Bubble.io's straightforward data management, building something like a simple CRM would force your users to cap their customer base at 10,000. Not ideal for a business hoping to, you know, grow.

Lesson one: Bubble.io’s database is like a budget airline — great for a quick trip, but you wouldn't want to rely on it for a world tour.

The quest for more robust data management

Given these data limitations and the potential for escalating costs as your data expands (thanks, work units!), many Bubble.io veterans suggest using external data storage. Think of it as renting extra baggage space for your budget airline flight. My choice was Supabase, which seemed like a solid solution — until it wasn't.

Problem #1: The integration. The most popular plugin for linking Bubble.io with Supabase allows for a maximum of five filters. Want more customization? Sure! But the options are as well-documented as the Loch Ness Monster — everyone talks about them, but no one can actually show you where they are. So, if you're thinking of going this route, be prepared for some serious trial and error.

At this point, you’re left with three not-so-great options:

  1. Bubble.io storage: Perfect for tiny, simple apps. Think “personal to-do list,” not “customer management system.”
  2. External storage: Great for handling larger datasets and more complicated models but often still limited by integration constraints and requiring significant logic in storage solutions. Kind of like plugging a Tesla battery into a skateboard.
  3. Web services and APIs: Moving all your data to a different solution and accessing it through web services. However, this isn't very low-code, is it? More like low-code adjacent.

UI and UX: the heart of any modern application

When I started playing with Bubble.io, I was hoping to build a user interface without getting bogged down in the complexities of modern JavaScript frameworks (which, to be honest, I’ve never been thrilled about). For the first few hours, Bubble.io felt like a dream come true. Creating UI elements was quick, easy, and surprisingly fun. But as soon as I tried to do anything beyond the basics, the dream started to crack.

Creating something as simple as a switch with custom icons turned out to be a mini-nightmare of workarounds and hacks. And in today’s world, where user experience is often the deciding factor in whether your application succeeds or fails, this basic approach doesn't quite cut it. Sure, you could write your own plugins (as many forums suggest), but that feels like a betrayal of the core promise of no-code — making development accessible and fast.

Bugs and other reasons to avoid Bubble.io for production

Now, let’s talk about the little quirks — or, to put it bluntly, the bugs. Some of these are more “lack of features” than actual bugs, but they’re still problematic. For example, you can’t “auto-bind” a checkbox. This issue was first reported back in 2020, and guess what? It’s still unresolved, and yes, I had to use a clunky workaround.

Even some of the plugins released by Bubble.io themselves are inconsistent. And then there’s the fact that the whole Bubble.io infrastructure is based in the US. While that’s fine for many users, it does create latency issues if your data sources and web services are located elsewhere in the world. Picture a digital tug-of-war with your data. Not fun.

Why Bubble.io still rocks for some projects

Despite these challenges, there are a lot of things to love about Bubble.io. It’s a fantastic tool for building basic applications incredibly quickly. You can knock out an MVP in record time, and it takes care of repetitive tasks — like user authentication — with ease. Integrating solutions like Stripe for subscriptions, payments, and invoicing is a breeze and can be done in mere minutes.

And let's be clear: Bubble.io is great for prototyping, creating mockups, and building simple apps. It’s just not designed to scale up to enterprise-level applications. But that doesn’t diminish its value for what it is designed to do — give creators and startups a fast and easy way to bring their ideas to life without drowning in code.

The verdict: fun, fast, but not fully-featured

Look, I'm not saying Bubble.io is bad. In fact, it's a fantastic tool for playing around, prototyping, or even building some simple applications. I’ll undoubtedly return to it for future fun projects. But would I stake an enterprise-level application on it? Probably not.

As experience shows, the real challenges start when you need to scale your applications, introduce complex functionalities, and resolve issues quickly. That's where Bubble.io's charm fades, and the limitations start to bite.

Can’t wait for Bubble.io 2.0

That said, I’m excited for the future. I can't wait for something like Bubble.io 2.0 — a platform that retains the ease of no-code but offers more flexibility in UI, logic, and data storage. Imagine a world where we could build powerful, scalable applications without needing to move logic elsewhere or juggle multiple external services. Now, that would be a game-changer.

So, if you’re in the mood for some no-code fun, Bubble.io is your friend. But if you're planning to conquer the world, you might need a bit more than just bubbles.

PS: If you’re still reading this and itching to try out something similar or just want to chat about no-code adventures, feel free to reach out. Who knows? Maybe we can build something fun together — like an app that orders a pint of Guinness for every new user. 🍺 Cheers!

PS2: Here is an application I created as my POC: https://stockmood.ai/

Konrad

IT manager, CTO, always passionate about software craftsmanship. Successfully delivered SaaS products from scratch to $17 mln ARR global solutions. Scaled IT team from 1 up to 75 people. Still know how to code ;)

Submit Your Comment