A response to Can Low-Code Platforms Really Handle Fully Custom Design? (November 2024)
In November 2024 I took on one of the most common objections web developers raised against Low-Code/No-Code platforms. They said page builders couldn’t handle truly novel design. I didn’t dismiss the objection. I conceded it:
“a structured Low-Code/No-Code framework, no matter how flexible, cannot possibly be as open-ended as a blank slate hand-coded site.”
Then I made my case on economics. A pure canvas site, built by hand, could do anything. But the more clever and unique the design, the more it would cost to design, to build, and to maintain over time. So I laid out the choice like this:
“Stick with hand-coded blank canvas development with extremely high costs and low flexibility”
or
“utilize a Low-Code/No-Code framework and spend a fraction of the cost while gaining massive flexibility, but recognizing that somewhere out there a designer may come up with a clever concept that might be just slightly out of reach.”
And I gave my answer: “I’m going with Low-Code/No-Code.”
That was the right call in 2024. WordPress plus Elementor really did handle nearly every design we threw at it, and it did so for a fraction of the cost of hand coding. I meant every word.
The trade off has disappeared
Look closely at how I framed that choice. One side had extremely high cost and low flexibility. The other had low cost and high flexibility, with a small ceiling on design. The whole argument rested on one assumption: that writing custom code is slow and expensive.
That assumption is gone.
AI agents now write and revise custom code under experienced direction. The blank canvas I described, the one only limited by the core technologies themselves, is no longer reserved for clients with very large budgets. We can build a design exactly as it was conceived, with nothing slightly out of reach, at a cost that is lower than the page builder approach I recommended.
So the choice I set up now looks like this. Blank canvas design freedom, at a lower cost, with lean code underneath. Or a page builder that costs more to run and still has a ceiling. I don’t have to think about this trade off for very long either.
What about flexibility?
In 2024 the biggest point in favor of low code was flexibility. Hand-coded sites were rigid because every change meant going back to a programmer, and that cost money. Page builders gave us the ability to try ideas and change our minds.
That flexibility now belongs to custom code too. When an agent can rebuild a section in minutes, changing course is no longer expensive. We can try one layout, look at it in the browser with the client, and try another, without the old programming overhead. That is the kind of experimentation I was always after, and we get it without carrying a builder’s worth of code for options the site will never use.
And for clients who want to edit their own pages, we include Simply Edit, our on-page editor, which understands the structure of the custom code. You get the freedom of a custom design and the ease of editing you were used to with a page builder.
The cost of maintaining something unique
I also argued that bespoke sites were expensive to maintain over time. That was true when every update required a developer to dig back into hand-written code. It is much less true now. The same agents that build the site can read it, update it, and extend it, with someone experienced reviewing the work. A focused custom feature is often easier to maintain than a large plugin and its add-ons, because there is simply less of it.
Designers get their vision back
The most important thing hasn’t changed. I have always wanted designers and strategists to be able to express their ideas on a working website without losing them in the handoff. Low code gave us that for a while, within limits. Agentic development gives it to us without the limits.
So the clever concept that might have been slightly out of reach in 2024 is now within reach. And it costs less to get there.
¡Viva la Revolución!