How concerned are you with scalability when using no-code tools to build software?
Felix
11 replies
Replies
Bakemono Fra@bakemono
Quite a bit. I think that if you want to build something that is truly scalable and potentially growing in complexity you should go the yes-code route. No-code is great but for specific use cases
Share
@felix_momen Webflow and Framer are very good in my opinion :)
@key_lu I would say when you start implementing features that go beyond very common use cases. I know some tools can be pretty powerful and able to do crazy stuff, however in my experience complexity spirals out of control pretty fast, and at that point I'd rather go with code: maximum flexibility, customisation, and control. But hey, I could change my mind soon, the tools are improving every day š This is just my opinion at this specific point in time. I also think that the future could be a tightly integrated hybrid of the two worlds. No-code as the basis for most stuff, code for more complex stuff. I see a lot of potential in this direction. Bonus point: coding is becoming easier and easier thanks to copilot and other AI tools so yeah the two worlds might merge more in the future
It is not simply about scalability; but rather about the steep, non-linear increase in costs that accompany scaling up
@jiangyaokai assume platform performance can be managed, costs zooms up fast
TheSprkl
I think it also depends on the tool itself. Some tools are good just for the MVP, to launch quickly and start growing. But some are pretty good for the scaling as well, for example, Webflow - you can build initially some promo page and then transfer it into the big website for your business / product.
My take is that this question should really be split into two sub question:
Is the platform scalable in terms of performance?
Can products built using a no-code platform be maintained and iterated upon?
Regarding performance: save a few giants like facebook / amazon, I would argue that 99.9% companies will never exceed 1m users. So if a platform can support a 1m user base, which I would say approximately translates to peak traffic of around 2 million API invocations per minute. Not that many no code platforms can support such traffic at this moment.
Regarding developer performance: The concern comes from two places: can I onboard a new developer quickly? Can a product be expanded beyond the platform's limits in terms of functionality?
I'm using Webflow to build an MVP. To go from zero to one, this is ideal for me. It's low cost and something I can mostly build myself. However, it's inevitable that custom software will be required eventually.