They pay every month, don’t open support tickets, don’t complain, don’t ask for anything, but if you look at the revenue sheet, there they are, contributing as much as anyone else.
They’re dormant customers, and in almost every company I’ve worked with, there are quite a few more of them than anyone wants to count.
It’s not that they’re angry. It’s something more uncomfortable: they don’t cancel because canceling takes work, too. They have to log in to the platform, find out where to cancel, tell the person who approved it in the first place, and explain why the thing they bought with such enthusiasm is no longer being used… And when your day is full of fires to put out, that task always gets pushed to tomorrow. It’s not urgent: “It’s just a small payment we can deal with later.” So they keep paying for a subscription they barely look at, month after month, out of inertia
But one day, your team finally has some time and decides to follow up with the accounts that have been quiet for a while.
You call.
And the customer remembers something…
The problem with calling when you finally have time
I’ve seen that call end in a cancellation enough times that it no longer surprises me.
Someone picks up, hears “How’s everything going?”, pauses for a second, and answers politely until, a few minutes later, they say something like, “Well, now that you mention it, there’s something I wanted to tell you.”
It’s easy to draw the wrong conclusion from that: you’d have been better off not calling.
But no, that’s not it.
That call didn’t cause the cancellation; it just put a date on it. That customer had been gone for months. They simply still had a recurring payment set up and were still paying.
What was wrong with that call was the timing, the day, not the way it was made.
The timing, because it came when the company had time, not when the customer gave the first sign, and there can be six months between those two moments. What’s more, asking “How’s everything going?” is the worst question you can ask someone who isn’t using your product; you hand them the agenda for the conversation and force them to admit out loud that they aren’t using it. Once it’s in their mind, once they’ve said it, they can’t take it back, because they know they should have done something about it long ago. What comes next is a consequence.
Not all silence means the same thing
Before going on, there’s an important distinction, because otherwise this turns into chasing everyone who doesn’t reply (we know Halloween is coming, but we don’t want a witch hunt 😜) , and that neither pays off nor makes sense.
So here are the different kinds of silence:
- Healthy silence. This is the customer who logs in every day, does what they need to do, and has nothing to ask you. You can recognize it because their usage is steady and boring: the same screens, the same frequency, month after month. They don’t write because everything works, and every time you call to “see how things are going,” you’re taking fifteen minutes of their time for nothing. With highly operational products, this describes most of your customer base, and mistaking it for a risk is an expensive error: you spend your team’s time on accounts that don’t need it.
- Inertial silence. This is the customer who logs in now and then, uses a small part of what they bought, and has stopped making progress. You can usually recognize it because they’re stuck on their first use case: they’re doing the same thing they were doing in their third month, never moved beyond it, and if they bought five modules, they use one. They’re not unhappy, and if you ask, they’ll say things are fine. But they’re paying for value they’re not getting, and that only lasts as long as nobody at their company takes a close look at the bill.
- Terminal silence. This is the customer who no longer logs in. The person who championed you changed roles and nobody told you, or they’re using an alternative tool you don’t know about, or the project they bought you for was canceled eight months ago. At this point, the account is alive only in your billing system.
You don’t see the difference between the three on the phone; you see it in usage. And if you’re not looking at usage, all three look exactly the same: quiet customers who cause no problems.
Why nobody looks after them
This is the structural part, and it explains why this happens even on teams that work incredibly hard.
In almost every company, whoever makes noise sets the day’s agenda. The customer who writes, who escalates an issue, who has an incident, who asks for training, who’s negotiating a renewal next week… all of that lands in the work queue on its own, without anyone having to decide anything. It’s what you can see without a system.
But the quiet customer doesn’t enter any queue, because they don’t generate a task. They don’t show up in email, they don’t show up in chat, they don’t show up in the Monday meeting… and what doesn’t show up simply doesn’t get worked on. It’s not that the team decides to ignore them; there’s no mechanism to put them in front of anyone.
Then there’s the other reason, the one almost nobody says out loud: as long as you don’t look, they still count as revenue (and yes, I’ve seen this at several companies). Looking means admitting that part of the customer base isn’t healthy, and that makes anyone who has to present the numbers uncomfortable. And because we need the money, we look the other way, at something else.
That’s why you don’t fix this by asking the team to pay closer attention; you fix it by making silence generate work, just as a complaint does. Until it does, until you find a system for actually reviewing your customer base and segmenting it, the one who shouts will always win.
What’s really happening with your numbers
Brace yourself, especially if you’re growing.
When you have a large pool of dormant customers and keep adding new customers on top, the dashboard says everything is fine. Revenue is rising, retention looks stable, and nobody has a reason to look underneath. But you’re not growing on a solid foundation; you’re growing on revenue you’ve already lost but haven’t been told about yet.
Do the math with round numbers:
- Say that out of every hundred customers, twenty haven’t logged in for ninety days.
- Those twenty are in your revenue figures and next year’s forecast. They’re in what you tell an investor and in the decision to hire two more people.
- And the probability that they’ll still be there in twelve months is considerably lower than your historical churn rate suggests, because your historical rate measures the customers who left, not the ones who have already left but are still paying.
And that bill comes due all at once, because on the day someone takes a serious look at the quiet accounts, or the day a customer reviews their spending, or the day a new manager comes in wanting to clean up the vendor list, it won’t be just one customer who leaves.
Several leave at once, and then it looks as though something strange happened that quarter. What’s really happening is that two years of accumulated losses are being realized.
That’s why, when people ask me whether anything can be done at that point, my answer is no.
At that point, you can’t fix it; you have to accept it. What you can plan for is everything that comes before.
The zombie almost always starts in onboarding
In my experience, dormant customers rarely fall asleep all at once. Usually, they never woke up in the first place.
They signed with great enthusiasm, implementation took longer than expected, you had to wait for their IT team to grant access, summer arrived, the person championing the project changed priorities, and by the time everything was ready, nobody remembered why they’d bought it. There was never a moment when that customer saw what you’d sold them working with their own data. And because they’d been paying from day one, no alarm went off.
That leads to my favorite measure, because it’s predictive rather than reactive: if you want to know how many dormant customers you’ll have a year from now, look at how many customers who signed up this month still haven’t used the thing they bought your product for. That number, not your churn rate, tells you what your customer base will look like next year.
So, can you plan for it?Yes, and you don’t need a new tool for any of these four things.
- Define what dormant means, using a number. Silence isn’t a signal; it’s the absence of one, which is why it triggers nothing: nobody has an alert saying “this customer hasn’t done anything.” If you don’t decide what it means, someone’s intuition will decide, and always too late. You can start with two data points: how many days it’s been since they logged in and how much of what they bought they’re using. That’s enough to distinguish the three kinds of silence.
- Act after thirty days, not six months. A customer who cools off in the first month can still be won back, because they still remember why they bought from you and haven’t yet reorganized their work without you. After six months, they’ve found another way to do things, and a call won’t reverse that: you have to sell to them again from scratch.
- Make silence generate a task. The format doesn’t matter; it could be a list someone reviews every Monday. What matters is that the mechanism exists and someone owns it, because if it depends on someone remembering, it won’t happen.
- And when you come back, bring something with you. Never “How’s everything going?” Bring a piece of their own data, something that’s changed in the product and addresses exactly what they came for, or what similar customers are achieving. Then the conversation stops being “Are you still interested?” and becomes “Look at this.” It won’t save all of them, not even close, but it changes the percentage considerably.
And a practical warning that almost nobody gives: if you have a lot of dormant accounts, don’t wake them all at once. Start with a small batch, see what happens, learn what works for them and what makes them hang up, and then continue.
If you call two hundred in two weeks without testing your approach first, all you’ll do is concentrate into one month a set of cancellations that would otherwise have been spread across the year, without learning anything along the way.
What I wouldn’t do
Three things I often see people do that cause more harm than good.
- Send an automated “we miss you” email. It reminds them they’re paying for something they don’t use, from a sender they don’t know, with the cancel button just two clicks away. If you’re going to remind them you exist, offer something useful and make sure there’s a person behind it.
- Offer a discount to get them to stay. A customer who doesn’t use the product won’t stay because they’re paying less. They’ll stay a couple more months and leave anyway, only now they’ll have paid you less and learned that your price was negotiable. A discount addresses a price objection, and this isn’t a price objection.
- Chase healthy silence. If the customer logs in every day and has what they need, let them work. Filling their calendar with check-ins to prove you’re supporting them is the best way to turn a happy customer into one who starts wondering how much they’re paying you.
And let the ones who don’t come back leave on good terms
This won’t bring them all back, and it’s worth saying so, because claiming otherwise would be selling you a fantasy.
Of the batch you wake up, some will cancel. That’s what will happen. But there are two decisions worth making before it happens, not while it’s happening.
- The first is whether you have something smaller to offer them, such as a lower-tier plan, fewer licenses, whatever it may be. Offer it, because a customer who downgrades is still a customer, still knows you, and may grow again when their situation changes. One who leaves entirely usually doesn’t come back.
- The second is to ask the right questions when they leave. Not with a five-star survey, but in a ten-minute conversation where your only aim is to understand the exact point when they stopped using it and what was happening at their company then. That information is worth more than the account itself, because it tells you where your product or delivery is failing for all the other customers who haven’t left yet. It gives you a head start on solving a problem they already have or will have.
Once you’ve done that, close the matter. A customer who leaves feeling well looked after will speak well of you. A customer you chase to the bitter end will speak badly of you, or won’t speak of you at all, and you’ll cease to exist for them.
To finish
There’s a test I like more than any health score for finding out whether an account is truly alive:
If you turned off the service tomorrow without warning, how long would it take them to call you?- If the answer is ten minutes, that customer is alive, even if they haven’t written to you in months.
- If the answer is that it might take weeks, or they might not call at all, that isn’t a quiet customer. It’s a customer who has already left and just hasn’t told you yet.
And this is exactly what we saw this week with ĀticcoLab and its students
Ask yourself that question about your twenty quietest accounts. How many would call?
