Software has become very eager to help.
Open an app and you're greeted with a product tour. Dismiss that and there's a prompt to enable notifications. A few minutes later you're asked to rate the app. Somewhere in the corner is a chat bubble waiting to answer questions you haven't asked. Before you leave, there's a reminder to subscribe to a newsletter.
Each interruption is easy to justify on its own.
Together, they make software feel exhausting.
Most of these features exist with good intentions. Product tours reduce confusion. Notifications bring people back. AI suggestions promise to save time. Feedback prompts help developers improve the product.
The problem isn't any single one of them.
It's that every feature assumes your attention is available.
Attention is one of the few things software can't ask for without taking it away from what you're trying to do.
One reason this happens is that it's easy for the people building a product to overestimate their product's importance in someone else's day.
After spending months or years building an app, every feature feels significant. Every announcement seems worth showing. Every improvement deserves attention.
For the person using it, though, your app is often just one small stop in a much busier day. They aren't thinking about your roadmap, your latest release, or your engagement metrics. They simply want to accomplish a task and move on.
Designing with that in mind leads to different decisions. You become more selective about when you interrupt, because you stop assuming your product deserves attention simply because it exists.
Helpful software understands this. It waits. It stays out of the way until there's a genuine reason to interrupt.
A spell checker highlights mistakes after you've written something. Navigation reroutes when traffic changes. A backup runs quietly in the background. These are helpful because they solve problems without demanding attention first.
Whenever I'm building something, I try to ask one simple question:
If we removed this interruption entirely, how many people would actually miss it?
Often, the answer is very few.
The best software isn't the software that offers the most help.
It's the software that knows when not to.