The Hidden Cost of No-Code
I'm a huge fan of no-code app builders, but I'm also a software developer and I've been the first tech hire at a couple of startups now. Here's how I look at no-code:
My standard advice is to start with something like Bubble if you need to build a web app just to launch. Bubble lets you push off building a real product for months or years, but, like any technical debt, it will drag you down if you grow fast enough.
So for funded startups, no-code doesn't always make sense. If you have $1 million in the bank, why cut corners?
What do you think?
Even if you have $1mil in the bank for your startup, nocode can still be great for:
Thanks for the sharing, and I thinks its a happy problem! I'm using Bubble on a new product too and I'm also a software engineer myself. Wondering if you can share more about the specifics of bottleneck at Bubble which you think is technical debt? And would love to learn more about the architecture too.
I use ec2 as backend and bubble has the frontend and DB with user data and payment data only, since our product is to let people use our APIs, those API directly call our ec2, and ec2 query the user data from Bubble for token verification. And I'm wondering if the Backend workflows at Bubble is good to handle larger volume of query in the long run...
I like the hybrid approach. That will probably make it easier to transition off Bubble if you eventually have to.
When I used it a few years ago, the startup started to have slowdowns once the app got big (20+ screens) and they were driving 10k+ weekly users.
If you need to reach millions of users, you'll eventually have to migrate off Bubble because the optimizations are pretty specialized, but if you can keep it low volume, you might be good for years to come.
My current business is all run on Zapier and Airtable, but we're a B2B service, so we don't have huge volume. At some point, a lot of no code solutions get slow or cost restrictive if you scale them way up.
Thanks for the sharing! And I agree that eventually it has to be migrated off if it's really going to scale up, can't fine tune performance if it's relied on external software afterall... and I look forward to this day when I have the happy problem too^^
Thanks for the post, and it's a good argument on the topic.
I would think No-Code is a good way to verify the business model or the demand for a solution, quickly. It's really quick to implement a MVP and test the market demand for your solution.
Once you have some signals, then perhaps it will be wise to move away from No-Code.
Hope that make sense and helps 🙏
If you have $1M in the bank of course you can afford developers to work for you. What if you have no even $100k ? Nocode tools are perfect solution for those who want to create something to test an idea, to get fast time to market, etc etc, and have no budget to hire pro developer... When company grows and user base is increasing obviously you will need to rebuild your product to match higher requirements. Some well known apps like i.e. Whatsapp (if remember correctly) , were rewritten more than once, and tech stack was evolving as growing load required to rebuild app , change architecture etc etc. I think it is rather rare that even in funded startups first choice of tech stack is optimal and lasts forever. Most often the choice is made based on what developers knows best - languages, tools etc That allows them to rapid build first version of product .
This comment was deleted 5 years ago