top of page

Are You Sure? Writing Confirmation Modals

I hate tomatoes. It's unfortunate, since it adds even more to the limitations of being a pescatarian. People can accept that I don't like meat or kefir. But my in-laws won't accept that I don't like tomatoes.

Are you sure? Well, yes. I know what I like and don't like. I have had tomatoes before. But my family insisted that the problem was I'd never had a good tomato before. So I tried it again: and still disliked the bitter taste. And again, and again.

I can understand the Are you sure if I'd never before had a tomato, and had just decided not to like them based on look alone. And while the question was asked out of genuine concern, it lost its charm every occasion it was asked.

If we know that repetitive questions come across as patronizing, non-listening, and unhelpful, why do we do the same thing to our users?

Don't worry: I'm not arguing against the confirmation modal as a whole. Error prevention is an important part of good design and writing, but we can do it in a better way.

Take a look at the examples below. They follow the formula of 1) asking if the user is sure they want to perform an action, and 2) offering a way out.

This isn't a case of horrible UX writing; if something like this was in production, I wouldn't immediately panic. But it's a wasted opportunity.

The entire modal exists only as a safeguard against mistakes. But how do I know if something is a mistake or not? Imagine if the subtitles here provided information that could help the user decide whether to continue or turn back. Something like:

Title: Delete file?

Subtitle: This can't be undone.

CTAs: Yes, Delete / Cancel

Again: nothing revolutionary here. But the modal now provides relevant information, a way of saying Hey, feel free to delete this, but there's no chance of getting it back. Confirmation modals shouldn't just ask are you sure; the real question is do you know the consequences:

  • Can the action be undone?

  • Will the action affect only me or others?

  • Will the action affect other parts of my experience?

There's nothing wrong with providing a moment for users to make sure they want to do what they're about to do. What is wrong is expecting them to make a decision without knowing what that means.


Thanks for subscribing

bottom of page