We as a society have thankfully moved on from the cloying, cringe practice of “confirmshaming.” But that is no reason to let your hard-earned revenue roll off a cliff.

This is the story of how I persuaded Discord Revenue to persuade our users not to cancel their subscriptions — without being scumbags about it.

Avoid the wrong problem

Discord is a freemium app with premium subscriptions under the “Nitro” brand. When I was the embedded content designer for the Nitro growth squad, Nitro upsells were everywhere in the product — but only for non-subscribers. Subscribers couldn’t tell a free feature from a paid one, which lowered the perceived value of their plan.

This was a system-wide design problem that our team was unfortunately not empowered nor expected to fix. And so, in the meantime, the team was attacking this problem from another direction: a post-churn survey.

But in the process of attempting to ask users why they were canceling, I found a problem.

We were doing nothing to stop them.

Sometimes… things that are easier… are worse.

I discovered that Nitro subscribers could cancel their subscription in 3 steps:

1. Open Settings › Subscriptions
2. Click “Cancel”
3. Click “Confirm”

So fast! So easy!

I hated it.

<aside> 🗣️ Hear me out…

1. It is business malpractice to allow your exit ramp to be a waterslide. All those resources, meetings, slide decks you spent on growth, acquisition, engagement? Wasted, because you forgot about basic churn prevention.
2. There is nothing deceptive about straightforward friction. In my view, it is very easy to stay well within ethical and legal bounds as you work to ensure a user’s intent to part ways with your product is well-considered rather than, well, capricious. Asking for additional confirmation before a destructive action is good UX!
3. Users deserve to know what they’re paying for — and what they stand to lose. The challenge is in finding the right time and place to educate them. SaaS products roll out new features all the time, but there are precious few organic moments when your users will be receptive to being “re-sold” on their existing subscription. One such moment is when you’re 100% sure they’re about to cancel it. </aside>

Make the case

I was adamant that we needed to insert some friction into the flow between “Cancel” and “Confirm.” I mocked up a wireframe that explored 2 added steps:

1. ‘Need Help?’ — This screen would redirect users to manage their payment method, choose a different subscription tier, or contact customer support. But this felt like overkill, and could come off as patronizing to second-guess a user’s clear intent, so I scrapped it.
2. ‘What You Lose’ — I presented this to the Revenue team, insisting that this would yield immediate results. My design partner Brittany Forks loved it, and PM Christina Zou quickly added it to the roadmap.

Show them ‘what you lose’

I thought it would be useful — appreciated, even — to remind users what they were giving up.