The feature that should have been a deletion

A few quarters ago I sat through a roadmap review where the most celebrated line item was a new comparison tool — three tabs, filterable, exportable. It shipped, and usage was fine. But the metric that actually moved that quarter was a different change entirely: we deleted a settings page that had accumulated nine toggles over two years, six of which fewer than 1% of users had ever touched.

Removing things doesn’t get a launch announcement. It doesn’t have a demo. But every option a product carries is a decision the team is asking someone else to make on every visit, and most of those decisions were made once, for one client, in one meeting, and never revisited.

The audit that led to that deletion was simple: for every configurable setting, we asked who changed it in the last twelve months, and whether the product broke in any meaningful way if the answer was “basically nobody.” Where the answer was genuinely nobody, we removed it and watched support tickets, not vanity metrics, for two weeks.

What’s worth noticing is why those settings existed at all. Almost none were requested by users. They were requested by a stakeholder anticipating an objection, or added defensively after a single complaint, or left over from a migration where nobody wanted to make the call to drop them. Each one was reasonable in isolation. Together, they were the reason new users needed a walkthrough just to get started.

A roadmap with a deletion on it looks unambitious in a planning meeting and looks exactly right three months later, when the product is easier to explain, easier to support, and — if anyone’s counting — easier to sell.