Make Them Say It to Your Face

Most companies let customers cancel through a button. That's a grave mistake if you're early stage and every customer matters.

The way I set things up, nobody can cancel without going through us. When someone wants to leave, they have to reach out. And when they do, I just say let's get on a call.

I don't have some forced system where the cancel button is hidden. It's simpler than that. If you've built a real relationship with your customers, the ones that matter will tell you before they leave. They won't just click a button. The person who owns the deal needs to have enough of a relationship with the customer that leaving feels like a conversation, not a transaction. For the accounts that matter, that's how it should work.

That's it. No "I'm sorry to hear that." No "can you tell me what went wrong over Slack." Just: let's get on a call without anything else. The reason this works is that most people will say yes. They feel like they owe you the conversation, especially if you've been responsive and helpful during the relationship. And once you're on a call, everything changes because you can actually do something about it.

Most recently, a customer messaged me on Slack saying they wanted to cancel. I said no worries I'll sort this out, and then immediately asked if they had time for a call that day. They said yes.

On the call, I learned the problem. They'd been pulling personal phone numbers through our platform and a lot of them were dead. So their cold calling wasn't converting and they decided the tool wasn't worth it. Fair enough on the surface. But when I dug in, it turned out they'd never tried using our Google Maps block to pull business phone numbers directly, which is what most of our customers doing similar outreach actually use. They'd been using the platform for a use case that wasn't the best fit and didn't know the better approach existed.

I think this is actually one of the most common reasons people churn from early stage products. It's not that the product can't do what they need. It's that they can't get the same results on their own that you showed them in the demo. You're a professional on your own platform. They're not. The gap between what you demo and what they experience alone is where most churn lives.

People almost never leave because your product is fundamentally bad. They leave because they got stuck on one approach, it didn't work, and they assumed that was the whole product.

So here's what I did. They had about 10 days left on their plan. I said since you're already paying for this time, let me personally set up workflows for you and try a different approach. I'll give you extra credits so you have enough to actually test this properly. If it works, great. If it still doesn't, we part ways and at least you leave with some value from the remaining time you already paid for.

There's a few things happening in that offer that I think matter.

First, I'm not asking them to spend more money on another month. The person on the other end has already decided this isn't worth what they're paying. Asking for more money at that moment is a guaranteed no. Instead I'm saying use what you already have, and let me make it better.

Second, I'm putting myself on the line personally. Not "our support team will help you" or "here's a help doc." I'm saying I will set this up for you. I will get on another call. I will be hands-on until this works or it doesn't. That changes the dynamic because now they feel like they have someone invested in their success, which is probably what was missing in the first place.

Third, I'm framing the default as leaving. I'm not trapping them. I said if it doesn't work we can end it. That actually makes people more willing to try because the pressure of commitment is gone. They're just using time they already paid for with someone who's going to help them do it right.

This matters because it purely depends on whether you actually have a solution to their problem. If the issue is something you can fix, like them using the platform wrong or not knowing about a feature, then getting them on a call and being hands-on is the right move. But if you don't have a genuine solution and you're just trying to buy another month, you're going to damage the relationship. Take the feedback, let them go gracefully, and improve the product. You can always re-engage six months down the line and say hey, we've fixed the things that didn't work for you, come take another look. But only if you didn't make it painful for them to leave the first time.

And on that call he actually said that if we get real value out of this, they'll renew.

When I do save an account, the work doesn't stop at the call. I check in way more frequently for the first two to three weeks. I redo the onboarding. I get their broader team involved and look for additional use cases beyond the one that almost made them leave. You're basically rebuilding the relationship from scratch, and if you treat it like the original sale is still intact you'll lose them again.

Even when this doesn't save the account, the call is still worth it. Our head of product joined that call and got direct feedback on what wasn't intuitive in the product, which was something we'd been hearing from other customers too. That's information you don't get from a cancellation form. You get it from a conversation where someone is being honest because they've already mentally checked out and have no reason to be polite about what didn't work.

These calls are actually more valuable for product feedback than anything else we do. The team takes the feedback from churning customers more seriously than feedback from happy ones, and for good reason. This is a person who was sold on the product once and is actively choosing to leave. That's a fundamentally different signal than someone who never bought in the first place. Whatever they're telling you is a confirmed deal breaker, not a nice-to-have improvement. It's the difference between someone choosing not to buy and someone choosing to stop buying. The second one tells you a lot more.

The pattern I've settled into is pretty simple. When someone wants to cancel: get them on a call immediately, don't let it sit. On the call, figure out what they were actually trying to do versus what they ended up doing. If there's a gap between those two things and you have a real solution, offer to personally close that gap within whatever time they have left on their plan. Give them free credits or resources so the barrier to trying again is zero. And bring your product person on the call because the feedback you'll get in these conversations is better than anything you'll get from customers who are happy and not thinking critically about what's wrong.

The whole thing takes maybe 30 minutes per churning customer. Some of them you'll save. The rest will give you the most honest feedback your company has ever received. Both of those are worth the call.