Why We Moved Away From WordPress So Early

Sebastián Vélez — Tue, 11 Aug 2026 01:40:52 +0000

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.

Why Andrés and I Started Developwave

Sebastián Vélez — Sat, 25 Jul 2026 18:45:49 +0000

I met Andrés at EAFIT, in a group project, and what stuck with me wasn't that he was smart — plenty of people in that program were smart. It was that he actually wanted to build things, not just pass the class and move on. That's rarer than it sounds. We kept working together after that mostly because we both had that itch.

By the time we graduated (I finished Systems Engineering in December 2024), we'd already been freelancing on and off for a couple of years — small sites, a few broken deadlines, one client who ghosted us halfway through a project. Nothing glamorous. But it was enough to notice a pattern: most of the businesses we talked to didn't actually need "more software." They needed someone to sit down, understand how they actually worked day to day, and build around that instead of around whatever framework was trendy that month.

So in January 2024, we made it official and started Developwave. Andrés leans backend, I lean frontend and client relationships — but that split is looser than it sounds. We both end up on calls explaining to a client why something slipped, and we both end up writing actual code most weeks. It's a two-person company; there's no version of this where either of us just does one thing.

We're based in Medellín but work with clients across Latin America and the US, mostly custom web platforms, some AI/automation work, the occasional "can you just fix what the last developer left us" project (there's always one of those).

I'm starting this blog mostly for selfish reasons — writing things down forces me to actually understand them. So this won't be polished company-blog content. More like: here's a bug that took me four hours to find, here's why we picked Next.js over X for a project, here's a mistake we made pricing a contract that I'd rather you not repeat.

Topics I expect to write about:

Building with Next.js, TypeScript, and Python on the backend
Software architecture decisions and the tradeoffs behind them
AI and automation, from the "actually useful" side rather than the hype side
AWS/serverless stuff we've fought with
Running a two-person software company without losing our minds

If any of that's useful to you, stick around. And if you want to see what we're actually building, it's at developwave.com.

Thanks for reading.

DEV Community: Sebastián Vélez

Why We Moved Away From WordPress So Early

Why Andrés and I Started Developwave