Someone asked me to remove a feature last month. Not change it. Not improve it. Remove it.

It sounds simple. Delete the code, update the UI, move on. Except it never works like that.

A stakeholder came to us and said they wanted something taken out of one of our applications. They were confident. It wasn’t being used, it was confusing, it needed to go.

The problem is, I work with multiple groups of users within the same application. Just because one stakeholder thinks something is redundant doesn’t mean it is. They’re seeing their slice of the picture, not the whole thing.

So we didn’t just remove it. We asked why.

When I started digging, the story changed. The feature wasn’t broken or unused. It was being used differently from how it was originally designed. The workflows around it had shifted, and the feature hadn’t kept up.

What looked like “nobody uses this” was actually “this no longer does what we need it to do.”

That’s not a removal. That’s a development conversation.

There’s a reason every feature exists. Sometimes it’s a good reason. Sometimes it’s historic. Sometimes the reason has expired and nobody noticed. But there’s always a reason, and understanding it matters more than acting on a gut feeling.

A feature that feels irrelevant today might have been critical six months ago. The context that made it important doesn’t disappear just because the stakeholder forgot about it.

Removing something carries risk. You lose institutional knowledge. You break muscle memory for users who’ve adapted to it. And worst of all, you might end up rebuilding something suspiciously similar in three months when someone remembers why it was there.

More often than not, the conversation starts with “can we remove this?” and ends with “can we make this work better?”

The feature wasn’t serving its purpose fully. Processes had changed, requirements had evolved, and the feature was still doing what it was originally asked to do, which was no longer the right thing.

That’s not a reason to delete it. That’s a reason to develop it.

Everyone celebrates launches. Nobody wants to be the person who kills something. And rightly so. The decision to remove is almost never as clean as it seems.

Most features don’t need removing. They need maintaining.

That’s a much less dramatic conversation, which is probably why it doesn’t happen often enough.