What to do about a customer who leaves

A customer tells you they are leaving. Maybe they reply to a renewal email with two lines. Maybe they just stop paying and you notice a month later. Either way, you now have a choice. You can record that they left, watch your churn figure tick up, and move on. Or you can record *why*, in their own words, while the answer still exists.

Most businesses do the first thing. It is faster, and the number looks like it is doing its job. But a churn number and a reason are two different kinds of information, and only one of them tells you what to change. Good customer software makes the second one easy to capture at the moment it happens; in 360REV the departure and its reason live on the same record, so the note does not drift away from the person it describes. That is the whole of what we will say about our own product until the end.

A number tells you how many, not why

Customer attrition is a well-defined idea. As the encyclopedia entry puts it, "Customer attrition, also known as customer churn, customer turnover, or customer defection, is the loss of clients or customers." [1] That definition is about counting. It measures an outcome. It is silent on cause.

This matters because a rate is a summary, and summaries discard detail by design. If twelve customers left last quarter and fifteen left this one, you know the trend got worse. You do not know whether the extra three left for the same reason as the first twelve, or for three entirely unrelated reasons. The number cannot tell you, because the number was never built to carry that.

So the churn figure is useful for one job: noticing that something changed. It is a smoke alarm. It is not a diagnosis. Treating it as a diagnosis is how businesses spend a year "fixing churn" by lowering a price that was never the problem.

The reason is a different kind of data

A reason is not a number. It is a short piece of text, attached to one person, captured close to the event. "Moved to a competitor that integrates with their accounting tool." "Business closed." "Could never get the reporting to work and gave up asking." "Price went up and they did not see the value."

Each of those points to a different action. The first is a product gap. The second is nothing you could have changed. The third is a support and onboarding failure. The fourth is a pricing-and-communication problem. If all four sit inside one churn percentage, they look identical. Pulled apart into reasons, they tell you four different stories, and three of them are things you can do something about.

This is the oldest idea in employment, borrowed. When a member of staff resigns, sensible organisations hold an exit interview — "a survey conducted with an individual who is separating from an organization or relationship" [2] — precisely because the person on their way out will tell you things the people who stay never will. A departing customer is the same opportunity. They have no reason to flatter you and every reason to be blunt. That candour has a short shelf life.

Record it at the moment it happens

The single most common failure is not asking the wrong question. It is asking too late, or not writing the answer down anywhere findable. A reason you remember in your head is not data. A reason you record three weeks later is a guess dressed as a fact.

So the habit is small and specific:

• When a customer leaves, write the reason in their record that day.
• Use their words, not your interpretation. "Said the app was too slow" is evidence. "Performance issues" is already a conclusion.
• Record who told you and when. A reason reported by the customer and a reason you assumed are not the same quality of information.
• Keep it on the customer's record, not in a separate churn spreadsheet that nobody opens.

This is the same discipline that makes any customer system worth keeping. It is covered more generally in what a customer record should contain, and the reason a customer left belongs there alongside everything else you know about them.

Ask one question, not ten

You will not get a ten-question survey answered by someone who has decided to leave. You might get one question answered. Make it the one that matters: *What made you decide to go?* Then stop. If they say more, let them. If they do not, one honest sentence beats ten boxes ticked at random. The guidance in writing to a customer who has not replied applies here too — a short, plain, low-pressure message gets answered more often than a formal one.

Categories you can count, plus words you can read

There is a trap on the other side. Some businesses, having learned to record reasons, immediately crush them back into a dropdown: Price, Product, Support, Other. Now the reason is countable again — and it has lost most of what made it useful.

The answer is to keep both. Record the free text first, always. Then, when you review, sort each departure into a small set of categories so you can count them. The text is the evidence; the category is the tally. Do not skip the text, because "Other" is where your most important lesson usually hides, and a category can never be re-expanded into the sentence it came from.

This gives you something a churn rate cannot: the ability to say *why* the number moved. "Churn rose two points, and all of the increase is in the Support category" is a sentence that starts a fix. The principle that a figure is only worth tracking when it changes a decision is the subject of numbers that change a decision, and a reason is what turns churn into one of those numbers.

The five whys, applied to one lost customer

Once you have the reason, there is a simple way to get past the surface. The five whys is "an iterative interrogative technique used to explore the cause-and-effect relationships underlying a particular problem." [3] You take the stated reason and ask why, a few times over.

• They left because the price went up. Why did that lose them? Because they did not think it was worth the new price. Why not? Because they were only using one feature. Why only one? Because they never finished setting up the rest. Why not? Because onboarding stalled in week two and nobody followed up.

The stated reason was price. The real reason was an abandoned onboarding. You would have cut the price and lost money, and the next customer would have left for exactly the same reason. The point of asking why is to stop fixing symptoms.

What to do with what you learn

Reasons are only worth collecting if you act on them. The rhythm is quarterly, not daily. Read every departure reason from the period. Group them. Find the one or two categories doing the most damage. Fix the cause behind those, and leave the rest alone.

Resist the urge to react to a single loud complaint. One customer who left angry is a data point, not a mandate. The value is in the pattern across many departures, which is why the record has to be consistent enough to compare. A reason written one way this month and another way next month cannot be grouped, and ungroupable notes are nearly as blind as having no notes at all.

A short word on honesty

The reason a customer gives is the reason they are willing to say, which is not always the whole truth. "Budget" is the polite version of several harder answers. Record what they said, but stay sceptical of a category that is suspiciously tidy. If "too expensive" dominates your list and your price has not moved, the real problem is that you have not shown the value, and that is a different fix.

Where 360REV fits

Briefly: because the customer record and the reason live in the same place, a departure is logged against the person, the free text stays readable, and the quarterly review can group and count without anyone rebuilding a spreadsheet. The habit is yours to keep. The software's only job is to make sure the reason you captured is still there, attached to the right person, when you go looking a quarter later. For the broader case that a customer system is a set of habits before it is a tool, see what a CRM is actually for.