A bittersweet goodbye: Why I’m moving on after 5 incredible years on Bubble

Entrepreneurship
1

I’m writing this with a lot of respect for the platform that started it all for me. I hope this post is received in the spirit it was written, as a genuine ‘thank you’ and a reflection on how much I’ve grown here over the last five years.

Dear Bubble Team and Community,

After five years of learning and three years of active development across various accounts, clients, and plans, I’ve decided it’s time for me to move on to the next chapter of my journey.

I’m writing this primarily to say a huge thank you to this incredible community. Specifically, I want to shout out @adamhholmes, @boston85719, and the many others who have saved me, and hundreds of others, countless hours of debugging.

When I started, I didn’t know a thing about JS, HTML, or CSS. Through Bubble, I’ve learned more than I ever dreamed possible: backend logic, API integrations, database architecture, and so much more. This platform was my gateway into the tech world.

However, with recent advancements in AI-assisted development, I’ve decided to transition to custom code (for those wondering: I’m not using Lovable, Base44, or any of those AI-wrapper platforms. In my opinion, they are overpriced and don’t offer the control I’m looking for. Please do your own research before going down that route…). For our current needs, we require full control over our environment, native iOS/Android development, and the specific performance scales that traditional coding offers. While I know the Bubble team is working hard on performance, the speed of development in modern AI-powered IDEs has become too significant for us to ignore.

This isn’t a “goodbye” fueled by negativity. I’ve invested over $2,400 in the Bubble ecosystem and have brought many friends and clients into the fold who continue to build here. I truly believe in the mission of Bubble but in my opinion that might need to start looking in to giving access to the code and implementing some kind of IDE to stay in the game.

Who knows, perhaps one day Bubble will bridge the gap with a dedicated IDE-style editor that offers the best of both worlds, and I’ll find my way back. Until then, thank you to the Bubble team, the community, and the YouTubers who helped change my life.

Much love and best of luck :heart:

18 Likes
2

Let me guess. Cline + Claude sonnet 4.5 doing the heavy lifting?

3 Likes
3

We maintain a broad and sophisticated tech stack, and we’ve recently started exploring Google Antigravity since they released. So far, its performance has been outstanding.

3 Likes
4

If Antigravity has been performing well for you, that’s great. Just to be clear, my original comment had nothing to do with your stack or how “broad” it is.

If you like Antigravity, you should still give Cline or Claude Code a try. They need a little more setup than Antigravity, but the power you get back is on a different level. If Antigravity impressed you, these will probably melt your face.

1 Like
5

Thank you so much @Orbit for taking the time to write this. So glad we were part of your builder journey and good luck with your projects in 2026 and beyond :waving_hand:

3 Likes
6

Using Claude code since the beginning, it is amazing. But still, antigravity with running multiple agents with opus … very nice experience thus far. What do you like so much more about Claude code in comparison to antigravity with opus? Have to say, I just started with Antigravity but it feels as if it will give me more effective power than just Claude code.

1 Like
7

Thank you for tagging me and I’m glad I could be a small part of your journey and wish you the best of luck moving forward.

3 Likes
8

Thank you for the post.

You said you’ve spent about $2400 over the course of your Bubble journey…that’s 30-some dollars a month.

I’d be interested in hearing how much you made…and, if you think by switching you’ll make more.

Or, is this simply a case where you think you will need to do less work to make what you made?

There are Bubble apps that have had funding rounds and raised millions, and some have thousands of users and are still going strong on Bubble.

I came from the opposite spectrum, where I was in code for 10 years (still am at work), and I found Bubble and decided to give it a go for an app idea I had for a few years.

I’m about to go live with my second app and have enjoyed building on Bubble.

I found that with Bubble, I can give up complete control for some sleep. I let their engineers and others deal with all the problems, regardless of the time of day, and just sit back and let my app run.

Anyway, would be interested in knowing if you profited from Bubble and think you can profit more from another platform?

10 Likes
9

Hey @senecadatabase, thanks for an interesting comment!

I made around $1,000 in total from Bubble by building for clients, but I noticed a pattern: clients eventually wanted full ownership of their code and database structure (often moving to Supabase). While you can use Supabase with Bubble, it’s not a workflow I enjoy.

Most of my early costs came from the learning curve and building MVPs. When I saw friends build the same app in one week that took me a month in Bubble, I realized I had to switch. It wasn’t about being “lazy”; it was about efficiency. Now, I use an MCP server to set up my Supabase structure and Edge Functions for the backend. This allows me to focus on design (Figma, logos, wireframes) while the app is being built, which has easily doubled my productivity. Plus, having full SQL access makes things like generating demo accounts much simpler.

To answer your question: yes, I’ve made double what I did with Bubble in just a few months using Cursor and Antigravity. So for me, the switch was definitely worth it.

The three biggest reasons for the move:

  1. Time & Efficiency: The speed of development is on another level.

  2. Absolute Freedom: No more plugin limitations. If I need a feature, I just implement it.

  3. Product Ownership: I knew I would eventually need to transition to custom code after the MVP. Given how rapidly AI and IDEs are evolving, making the move now was the obvious choice. The biggest dealbreaker with Bubble was the lack of portability; you can’t truly take your work with you. While you can export data, the database structure is proprietary and doesn’t map easily to standard relational databases like Supabase. I wanted to build on a foundation I actually own and that is built for scale with full control.

However, it’s not all a bed of roses. There are significant downsides to leaving Bubble:

Security is the big one. You really need to know what you’re doing, or you will expose API keys and databases through poorly configured Row Level Security (RLS) or policies. If you aren’t comfortable with this, stay with Bubble. Accidentally exposing client data can end a career. Also, you have to handle your own rate limiting; otherwise, a simple script from a malicious user could result in a massive bill from Google or OpenAI.

I hope this answered your question :slight_smile:

2 Likes
10

Thanks for your response.

I appreciate you taking the time to explain your thinking.

From what you described, it doesn’t sound like Bubble was ever really stress-tested under scale, revenue pressure, or sustained production load. The switch seems to have been more by newer platforms and the ownership idea than by hitting limits.

That’s not a criticism, just a different optimization target. Some optimize for control and technical flexibility, and others optimize for speed to market and revenue. Right now, I’m much more focused on the latter.

With the clients I’ve worked with, owning the code hasn’t been that important. They seem to care more about owning the product, the users, and the revenue, and being able to iterate quickly without managing infrastructure or security at a low level.

If I ever reach a point where Bubble becomes a financial or technical bottleneck, that’s a great problem to have. Until then, the leverage Bubble provides is a big part of why I’m sticking with it.

3 Likes
11

Exactly, it depends on the use case. However, I’ve lost over 90% of potential clients because I couldn’t offer code ownership. When they asked, ‘Can we host this ourselves or expand production later?’ and my answer was ‘You own the data, but not the code,’ the deal died right there. I don’t blame them; they don’t want to be dependent on a single ‘Bubble developer’ when traditional developers are far more common and accessible.

2 Likes
12

Did you explain to your clients that owning the code and hosting it could cost several thousand dollars a month to maintain it?

And if they owned the code and hosted it, who would come to their rescue when things went wrong…which they always do?

Most clients need a little wake-up call on how much it actually costs to run a server, etc.

3 Likes
13

I didn’t tell them that because, in many cases, it’s simply not true. Self-hosting on a provider like Hetzner (or even for free on Cloudflare Pages) and using S3 or R2 for file storage is often significantly cheaper than Bubble’s pricing. Bubble acts as a middleman for services like Amazon S3, and they naturally have to charge a premium to cover the development costs of their editor.

As for who would come to their rescue: the talent pool for traditional web development is easily 100 times larger than the niche Bubble community. It is much easier to find a developer to maintain a standard codebase than it is to find a specialist for a proprietary platform.

2 Likes
14

It’s not any more expensive than bubble with WU, doing something as simple as adding the ability to create branches in bubble puts you at $209/m.

I have 8 instances hosted on Render which we use as part of our bubble app and it costs us a little under $200/m. Even for the DB, you can use supabase which is free and managed.

Of course it can technically cost thousands of dollars a month, our bubble bill alone is close to $900 with our plan plus WU overages.

2 Likes
15

To answer both questions…

I may be looking at this through a different lens.

In my experience, I’ve dealt with apps that are operating at a scale where monthly costs are in the tens or even hundreds of thousands of dollars a month. In that context, Bubble doesn’t really register as “expensive” to me, especially considering everything it includes.

Bubble isn’t just a server bill. It’s the editor, workflows, auth, scaling, security defaults, backups, and a platform that lets you build and iterate without managing infrastructure. When you look at it that way, the pricing feels pretty reasonable.

Raw infrastructure can absolutely be cheaper, and for teams that want to manage all of that themselves and are operating a little app, that makes sense. For me, I’m optimizing for predictability, speed to market, and focusing on the product and revenue, and Bubble has fit that well.

If 900 a month is stretching your budget, I would say other things are going wrong in your app that need to be looked at.

5 Likes
16

I think that’s like comparing a bicycle to a sports car. While both can get you from A to B, their structure, speed, and purpose are entirely different. You are referring to apps with monthly overheads in the six figures systems that likely support millions of users and generate massive revenue just to sustain themselves. Their infrastructure isn’t even in the same league as Bubble’s, making it a bit of an ‘apples to oranges’ comparison.

If Bubble charged “tens or even hundreds of thousands of dollars a month”, it would lose its entire user base. The platform’s core value proposition has always been providing a cost-effective way to build MVPs quickly for none coders, offering a much cheaper alternative to hiring developers or spending years learning to code yourself.

However, with modern AI-powered IDEs, it’s now even more cost-effective to spin up a Hetzner server and connect it to Supabase, R2, or S3. You can essentially use an LLM to architect your database and build out the entire application for a fraction of the price.

Furthermore, the massive overhead these established companies face is often a byproduct of their age. They are typically stuck with legacy systems and inefficient frameworks that drive up costs. Because these apps have been touched by hundreds of different developers over the years, they often lack a cohesive, optimized structure, making them far more expensive to maintain than a modern build, also a big reason why many of them are struggling to keep up with all the startups coming out of nowhere from people who started off as “vibe coders”

1 Like
17

The main problem is, you have analysis paralysis…

You’re talking about Bob, a business owner spinning up a Hetzner server and connecting it to a Supabase, R2, or S3…

this isn’t how things work. How many business owners do you know that could do this or want to do this?

There are those who are out to make money, and there are the hobbyists.

I’m out to make money. I have 6 successful businesses, plus own rentals.

I don’t look at every shiny thing that comes along.

Bubble works for me, and I plan on using it to make money.

A hobbyist really doesn’t make much money, but is swayed by every new shiny thing.

I had asked you how much you had made from Bubble and if you thought switching would make you more…

that was my main question.

You said you made $1000.

Ok.

I would have been more impressed with your story if you had said you hit some type of limit…

even though that would have been hard since there are apps on Bubble that have gotten millions in funding, etc.

So, I appreciate your decisions and hope you the best

3 Likes
18

I think in todays era the question is more fundamental. It is are you a skeptic or not.

These are not words of a skeptic.

19

I’m not a skeptic.

They made $200 a year for the last 5 years…

and now they’re announcing they’ve spread their wings and are flying away.

There have been people who have left before…

none have come back and announced the thousands they were making since leaving.

Let’s get real. You have everything you need right here in Bubble to make a fortune :slightly_smiling_face:

6 Likes
20

Doesn’t skeptic mean questioning things ? I used it as a good thing.

1 Like