What separates a SaaS business with 1% monthly churn from one with 3%? Three times the MRR after five years. No wonder everyone wants to bring churn down. The catch is that most teams go looking for a single magic formula, when a customer who leaves for lack of ROI has nothing in common with one whose champion has just resigned. In this article, we break down the main reasons for churn, how to spot each one early, and the action plan to trigger when it happens.
Lincoln Murphy, one of the best-known voices in Customer Success, boils churn down to two causes: either the customer went out of business, or they didn't get the outcome they were hoping for. It's a useful way to cut the topic down to size. But let's be honest, it's also a little thin when you're trying to decide what to do on a Monday morning. Behind every churned account lies a very different story: a customer who no longer sees the value, a champion who moves on, a product that falls short. Same symptom, radically different causes. And therefore different responses.
That's the whole point: stop treating 'churn' as one homogeneous block and tackle it reason by reason. Treat it as a single problem and you're guaranteed to deliver the wrong answer at the wrong time.
Then there's the question of scale. Churn compounds, and that's where it really hurts. Take a company growing 10% a month: at 1% churn, it reaches close to 43,000 in ARR after five years; at 3%, it stalls at 13,000. A ratio of one to three, for a two-point gap. Apply that to a real installed base and you're talking tens of millions in lost ARR over time. The order of magnitude is dizzying. In other words, two extra points of churn are not a detail. They're a slice of your growth that disappears every single year.
And yet, in the vast majority of cases, the departure could be seen coming long before it was announced. Falling usage, a contact going quiet, tickets piling up: the signals were there. You just need to know which ones to watch. Below, we walk through the eight main reasons for churn, how to anticipate each one, and what to do when it happens.
Reason 1: An ROI problem or weak perceived value
This is the number one reason, by a long way. The customer can no longer see what your product does for them. Sometimes they never really saw it; sometimes the early value has faded over time. Either way, the outcome is the same: when the invoice lands and the perceived benefit doesn't match it, the question is no longer 'will they leave?' but 'when?'.
Why it matters
A customer who doesn't see value won't defend your budget line at renewal. And watch out for a classic trap: satisfaction is not engagement. A customer can tell you they're 'happy' and still leave. Satisfaction measures whether the product meets expectations; engagement measures how indispensable it has become to their day-to-day work. They are not the same thing, which is also why NPS is a poor indicator of customer loyalty. A satisfied but disengaged customer is one who leaves without warning.
How to anticipate it
The signals are mostly behavioural: declining usage, adoption that plateaus on two or three features, a number of active users that stalls. These are typically the warning signs of disengagement that a customer never voices out loud. But the real safeguard needs to be in place much earlier: from the very start, agree with the customer what 'success' means for them, and measure it. Without shared objectives, you have no objective way of proving ROI when renewal comes round. If all you can do is hope they 'feel' the value, it's a coin toss.
What to do when it happens
Take back control with an outcome-focused Business Review. Not a demo of your latest product features, but a quantified account of what the customer has gained since signing. Reconnect their usage to their business objectives, identify the use cases they're not yet exploiting, and rebuild a short-term action plan with visible milestones. To structure the exercise, see our advice and best practices for Business Reviews. The goal fits in one sentence: prove again, in concrete terms, that the product delivers more than it costs.
Reason 2: A key contact leaves (losing your champion)
Your main point of contact moves on. The person who bought, who drove the project internally, who stood up for you in meetings you weren't in. They change role, leave the company, or get swept away by a reorganisation. And overnight, there's no one left to speak up for you internally.
Why it's a churn signal
Without an active champion, your product slides back to the margins. The replacement didn't sign with you, has no attachment to your solution, and often arrives with their own habits, sometimes even their own preferred tools. A new champion won't magically appear on their own. And an account that hinges on a single person is an account living on borrowed time, even when everything else looks fine.
How to anticipate it
Two reflexes. The first is multi-threading: never depend on a single contact. An account connected to three or more stakeholders can absorb a departure without wobbling. The second is to watch for the weak signals of a contact who is disengaging before they actually leave: slower replies, meetings pushed back, a cooler tone. These signals often come several weeks before the announcement.
What to do when it happens
Move quickly, because the first few weeks are decisive. Reach out to the replacement before they form an opinion on their own. Give them a mini-onboarding, as if they were a new customer, and above all, sell them the value again: never assume they know the history of the account. Reshare the results achieved, the current objectives and the success plan. In short, rebuild a champion rather than betting on the ghost of the last one. And keep track of your former champion: there may be an opportunity for you at their new company.
Reason 3: A 'bad fit' customer
This customer should never have signed. Their need doesn't really match what your product does, their size or maturity isn't right, or they bought on a misunderstanding. The misalignment was there from day one, hidden by the enthusiasm of the salesperson who closed the deal!
Why it matters
A bad fit almost always churns, and often quickly. No amount of support, however good, can make up for a product that doesn't solve the customer's real problem. Worse, these accounts eat up a disproportionate share of CSM time for next to no result, and skew your metrics along the way. The root cause almost always sits upstream, in a misalignment between Sales and CS: the deal gets signed for the number and the fit gets overlooked. You pay for it later. Every time.
How to anticipate it
The battle is fought before the contract is signed. A clear ICP, honest qualification during discovery, and a Sales-to-CS handover that flags warning signs from the outset rather than burying them. In the data, a bad fit shows up as an onboarding that stalls and a first result that never arrives. When both of those lights turn red very early on, it's usually not a support problem. It's a casting problem.
What to do when it happens
Be clear-eyed: it's the best thing for everyone. If there's a scope where the customer can get real value, reset expectations and focus them on it. But if the fit is structurally wrong, persisting costs more than it brings in, for you and for them. A clean offboarding that protects the relationship and your reputation is sometimes better than a battle you were always going to lose. Above all, feed the information back. Every bad fit is a data point that helps sharpen your targeting and avoid the next one.
Reason 4: A product-related problem
Recurring bugs, instability, an interface that's too complex, a critical feature missing, a roadmap that isn't moving at the promised pace. Here, the customer isn't struggling with the relationship or the value: they're struggling with the product itself.
Why it's a churn signal
A frustrating product pushes customers out faster than a price increase. Every bug, every missing feature, every slowdown chips away a little more at their trust. And it's the CSM who ends up on the front line, defending things they don't control and absorbing dissatisfaction they're not responsible for.
How to anticipate it
Keep an eye on the volume and nature of support tickets, the feature requests that keep coming back, and negative mentions in conversations. An account that's piling up product irritants is preparing to leave, even if it hasn't said so yet. The key is the link with the Product team: the CSM is best placed to report friction from the field, provided that channel is properly organised rather than left to chance.
What to do when it happens
Be transparent and proactive. Don't play the problem down and don't drown it in justifications. Acknowledge the friction, give visibility on what will be fixed and when, and offer practical workarounds in the meantime. In parallel, escalate the issue internally with facts, not feelings: 'three accounts worth €Xk each are blocked on this feature' carries far more weight than 'customers are complaining'. That's what turns a complaint into a roadmap priority.
Reason 5: Moving to a competitor
The customer leaves you for a competitor. A better price, a feature you don't have, more appealing positioning, or simply stronger sales execution on the other side. On paper, it's the most galling reason. In reality, it's rarely the true cause.
Why it's a false reason
If a competitor manages to dislodge your customer, it's almost always because a crack was already there: value that never took root, a relationship that was too thin, dissatisfaction that had been simmering. The competitor simply pressed where it hurt. One thing to watch closely: when you keep losing to the same player, it's no longer an isolated accident. It's a product or positioning signal to address at company level, not account by account.
How to anticipate it
Look out for competitors being mentioned in conversations, questions about features you don't offer, or a sudden interest in your exit terms. But the best defence starts much earlier: a customer who is firmly embedded, gets clear value from your product and has a relationship with you at several levels is far harder to poach. One useful principle: focus your retention efforts on the customers who have the most to lose by leaving, those for whom switching would carry a real cost in time, money or risk.
What to do when it happens
First understand the real reason before you react. Price, feature, relationship? Don't rush into a discount war out of reflex: a badly placed discount devalues your product without fixing the underlying issue. Restate what you bring that no one else does, bring in someone more senior if the stakes justify it, and deal with the real objection rather than the symptom. And if the customer leaves anyway, treat it as a learning case: a well-analysed loss to a competitor feeds directly into your product and your marketing.
Reason 6: The customer's financial health
The customer is in financial difficulty. Budget cuts, a cost-saving plan, headcount reductions, tight cash flow. Your product isn't being called into question: it's simply caught up in the trade-offs.
Why it's a risk
In a cost-saving plan, the first things to go are those seen as 'nice to have'. If your product isn't firmly in the essential tools category, it becomes an easy line to cut. Alongside this, there's a share of churn that's entirely involuntary (an expired card, a failed payment). It has nothing to do with the customer's health, but it can be fixed, and it's often revenue you can recover in full.
How to anticipate it
Pay attention to the customer's context, not just their usage: economic signals in their sector, hiring freezes, reorganisations, company news. An account whose market is struggling deserves closer attention well before renewal comes due. For involuntary churn, prevention is purely operational: automatic reminders before a card expires, monitoring of failed payments, and a follow-up process when a direct debit bounces.
What to do when it happens
Reposition your product as something that makes or saves the customer money. The famous 'must-have'. Quantify what the customer would actually lose by leaving. And if the budget constraint is real but temporary, it's often better to negotiate a downgrade, a pause or a plan adjustment than to take a clean churn: you protect the relationship and keep the door open for later. A customer retained at 60% of their contract is worth more than a customer lost at 100%.
Reason 7: Acquisition by a company that uses a competing solution
Your customer gets acquired. Good news for them, far less so for you when you discover the new group uses a competing solution. Overnight, your product is no longer being compared with the alternatives on the market. It's being compared with the acquirer's tool: already deployed, already paid for and already familiar to its IT department.
Why it's a reason in its own right
Unlike a move to a competitor, the customer doesn't necessarily have any complaint about you. They may be satisfied, have adopted your product in depth, and still leave. Because in a post-acquisition integration, the cost of change and the drive to harmonise tools often outweigh the comparative value of the two solutions. Rationalising the stack, consolidating contracts, having only one tool to maintain and secure: on paper, the argument stands up. Yet it can lead to a decision that makes no sense at all from the users' point of view. Politics and power games are never far away.
This scenario is more common than people think, and it's far from a lost cause.
How to anticipate it
Keep track of news about your accounts: acquisition announcements, merger rumours, a player consolidating its sector, a change of shareholder. As soon as the news breaks, map out the acquirer. Which solution does it use, at what scale, since when, and is it genuinely happy with it? Above all, who will make the call? The business sponsor you know well will no longer be the only one at the table. The real decision-makers are often elsewhere: in IT, which wants to reduce the number of tools to integrate and secure, and in Finance or Procurement, which want to consolidate contracts. Fail to identify them and the decision gets made without you.
What to do when it happens
Don't rush: play for time. Integrating two companies takes months, and migrating any given tool is rarely at the top of the group's priorities. Work with your sponsor and your champion to hold your position through the integration: secure the current renewal, document the results achieved, and put a figure on what switching would really cost (migration, training, lost functionality, the productivity dip during the transition).
Next, go after the group's decision-makers together rather than waiting for them to come to you. For IT, prepare your answers on security, integration and compliance. For Finance, bring a clear costing and a pricing proposal designed at group level. And offer a like-for-like comparison, results against results.
This is where the tide can turn. If the acquirer's teams are less happy with their tool than your customer is with yours, the question flips: why not roll your solution out across the whole group? Handled well, this churn risk can become a significant expansion opportunity!
Reason 8: The customer adopts an all-in-one platform that covers your scope
You're a best-of-breed solution, highly specialised in your field. Then one morning, your customer tells you they've just signed with a broader platform (ERP, all-in-one CRM, collaboration suite, and so on), one of whose modules covers, on paper at least, what your product does. The module is 'included'. Hard to compete with free... or so it seems.
Why it's an insidious risk
The generalist's argument is rarely functional. It's organisational and financial: one supplier, one interface, one contract, all the data in one place. Against that, the depth of a specialist tool carries little weight in an executive committee that thinks in terms of scope covered. The real trap is that this kind of project is decided far away from the users, and long in advance. By the time you hear about it, the contract has often already been signed.
How to anticipate it
In principle, this kind of project doesn't come out of nowhere. A tender, an overhaul of the information system, new leadership keen to 'simplify the stack': a champion you work well with will warn you well ahead of time, often months before the decision. Except in one case: when your champion is sponsoring the new project themselves and manages to keep it under wraps. Then you won't see it coming.
Hence, once again, the importance of never relying on a single contact. And of watching for the weak signals that give away a project in progress:
- unusual questions about data export or reversibility;
- a sudden interest in your termination terms or contract end date;
- questions about your connectors with a particular all-in-one vendor;
- a champion who pulls back from roadmap or usage expansion discussions;
- job ads from the customer that mention expertise in the all-in-one tool.
What to do when it happens
Don't fight on scope; fight on depth. In practice, a module within an all-in-one suite rarely covers every use case of a specialist solution. Pin down exactly what your users would lose (features, automations, time saved) and put a number on it. Day-to-day users are your best allies: they're the ones who feel the gap, not the steering committee.
Then propose coexistence rather than confrontation. Positioning your product as the specialist complement to the platform, connected to it, is often far more persuasive than fighting it head-on. You go from competitor to building block of the ecosystem.
And if the customer leaves anyway, handle the exit with care: a clean offboarding, a relationship preserved with the users, and a check-in scheduled six or twelve months later. Because the 'included' module doesn't always live up to its promises, and when the customer realises that, you want to be the one they're keen to call back.
The common thread: stop reacting, start anticipating
Eight reasons, eight action plans. But if you take away only one thing, let it be this: in almost every case, the departure could be seen coming before it was announced. Falling usage, a contact going quiet, tickets piling up, an acquisition announced. The signals are usually right in front of us. The problem isn't that they're missing; it's that they go unnoticed when a CSM is tracking dozens, sometimes hundreds, of accounts by hand.
That's where a health score changes the game. Built properly, it brings together the signals behind each reason (adoption, contact engagement, the sentiment of your conversations, tickets, deadlines) into a single view of risk, account by account. It turns scattered hunches into clear priorities. It's the building block that takes Customer Success from firefighting to fire prevention.
That's exactly the thinking behind Skalin. The health score calibrates itself from your real churn signals and detects risk several weeks before renewal, while there's still time to act. And because detection alone isn't enough, playbooks trigger automatically based on each account's signals: a nudge when usage drops, an outreach when a contact goes quiet, Business Review preparation ahead of a renewal date. Every reason has its signal. Every signal needs its response.
Churn will never fall to zero, and that's not the goal. But the difference between a team that suffers churn and a team that controls it almost never comes down to effort. It comes down to one thing: seeing it coming, and acting at the right time, on the right account, for the right reason.
Further reading:
