Vibe Coding Inside Companies Doesn't Pay

1 min read

Why do I think vibe coding inside companies is a bad idea? The answer is mathematical and surprisingly simple. ROI.

The setup

Let's say a SaaS subscription you need costs $200 a month. Even $1,000.

A developer comes to you and says he wants to build an alternative. It's free. He gives it to Claude and it works. That developer costs you about $100 an hour, and that's the good case.

What actually happens

He builds a first version in three hours. Another hour to get it live, because servers, databases, the usual. It works, but of course features are missing.

Now every week he spends another hour or two on tweaks and fixes, because let's be honest, it's fun. It's much more fun than the ticket he was supposed to be working on.

He also loses focus, because context switching has a cost too. An hour of tinkering doesn't cost an hour. It costs the hour plus the time it takes to get back into the real work afterwards.

And then there are the tokens. A few tens to a few hundreds of dollars more, in the good case.

The math

Add it up. In the end it costs you at least $1,000 a month in developer time and tokens (and that's if you got off cheap), instead of $200 for a system that somebody else maintains and takes care of, and which is probably a lot more stable.

Someone else has a team whose whole job is that one product. They handle the security patches, the edge cases, the 3am outage. Your internal version has one developer who is supposed to be doing something else.

It's fun. It's nice. But it's probably not economical.


All posts

Working on something this applies to?

A free 30-minute intro call. You'll leave with a clear read of your product challenges, ideas how to solve them, and a sense of whether we want to work together.