When Andrés and I started building websites, WordPress was one of the first things we tried.
It made sense. There were themes, plugins for almost everything, and you could get a website running pretty quickly.
But we didn't stick with it for very long.
The problem wasn't that WordPress was bad. It was that we quickly realized how complicated a "simple" website could become.
You'd install a plugin for one thing, another plugin for something else, then discover that two of them didn't really like each other. Updates could break things. Customizing something beyond what the theme allowed could get annoying very quickly.
And then there was the part we liked the least: when something went wrong, figuring out exactly what was causing it wasn't always straightforward.
We were still learning at the time, so instead of spending our time learning how to manage an increasingly complicated WordPress setup, we decided to learn how to build the websites ourselves.
That decision changed a lot for us.
We started building fully custom websites with React.
At first, that meant figuring out things that WordPress had hidden from us: how to structure a project, how to handle forms, how to work with APIs, how to deploy an application, how databases actually fit into everything, and what happens when you have to maintain the thing you built six months later.
We broke things.
We rebuilt things.
We learned a lot.
Eventually, we moved to Next.js for most of our projects, and our stack kept growing from there. Depending on the project, we've worked with things like AWS, Supabase, PostgreSQL, Python, APIs, authentication services, email providers, and various third-party integrations.
The interesting part is that none of that was really planned.
We didn't sit down at the beginning and decide, "We're going to learn all of these technologies."
A client needed something.
We didn't know how to build it.
So we learned.
Then the next project needed something different.
We learned that too.
That's probably been one of the biggest advantages of building custom software: every project forces you to understand something a little deeper.
And we've also learned that custom doesn't automatically mean better.
If someone needs a simple blog or a basic website they want to manage themselves, WordPress can still be a perfectly good choice.
But when a business needs something specific, we'd rather have the flexibility to build around the business instead of forcing the business into whatever a plugin or theme happens to support.
That's a big part of how we approach projects at Developwave today.
We're not attached to one framework or one technology. The stack depends on the problem.
For most of our web projects, that currently means Next.js and TypeScript, with whatever backend services and infrastructure the project actually needs.
Looking back, I'm glad we moved away from WordPress when we did.
Not because WordPress is obsolete.
Mostly because being forced to figure things out ourselves taught us much more than knowing which plugin to install.
And we're still learning.
That's probably what I'll write about here most of the time: the things we're building, the things that break, and the things we learn while trying to build better software.
If you want to see what we're working on, you can find Developwave at developwave.com.
Top comments (0)