The customer said they left for a better price. The churn report records it as price sensitivity. Neither captures the failed installation, the eleven-day delay, the service that never quite worked properly. The leave reason and the underlying cause are different things.
The churn report says the customer left for a better price. That’s what they told the retention agent when they called to cancel. It’s what the cancellation record says. It’s what the churn analysis shows — “price sensitivity” as the primary leave reason, consistent with the industry pattern.
What the churn report doesn’t say is that this customer had a failed installation, waited eleven days for the rescheduled visit, received a goodwill credit as an apology and spent the rest of their contract on a product that was provisioned slightly incorrectly and never quite performed the way it was supposed to. When the renewal came up and a competitor offered them a marginally better price, they didn’t need much persuading.
The leave reason was price. The underlying cause was provisioning.
Why customers lie on exit surveys
When a customer calls to cancel, they’re not trying to give you actionable operational data. They’re trying to cancel. The path of least resistance is to cite price — it’s simple, it’s defensible and it doesn’t require them to explain that the installation was a mess, the support agent couldn’t find their record and the service never quite worked properly for the first three months. Price is clean. The real story is complicated.
Retention agents are incentivised to match the stated reason with a retention offer — a discount, a loyalty credit, an upgraded package. They’re not incentivised to probe for the actual operational history and document the service failure that contributed to the customer’s willingness to leave. So the “price” reason goes on the record, the retention attempt fails or succeeds on commercial terms and the provisioning failure that contributed to the churn is never captured.
What the correlation looks like when you look for it
Operators who have done this analysis — matching churn records against operational history for the churned customer — consistently find a higher churn rate among customers who experienced provisioning problems than among those who didn’t. The effect is not immediate. A customer who had a rough installation doesn’t cancel the next day. The effect is at contract renewal — six months, twelve months, twenty-four months later — when the accumulated sense that the relationship hasn’t been great translates into lower switching resistance.
The numbers vary by operator and by the severity of the provisioning issues, but the pattern is consistent. A customer with a failed first installation is meaningfully more likely to leave at renewal than a customer whose installation went smoothly. A customer with a billing error in their first three months is more likely to leave than one without. The service experience in the first ninety days sets a baseline expectation that affects the renewal decision long after the specific incident is forgotten.
The measurement gap
Most operators don’t connect provisioning quality metrics to churn metrics, because they’re managed by different teams with different reporting lines and different KPI frameworks. The provisioning team measures installations, error rates and SLA compliance. The commercial team measures churn, retention and NPS. These data sets sit in different systems. Nobody routinely joins them.
When they are joined — even informally, even on a sample basis — the relationship between early service quality and later retention is usually visible enough to be commercially significant. The provisioning error that seemed to cost £150 to fix actually cost considerably more when its contribution to subsequent churn is factored in.
The implication for where to invest
If provisioning quality is a meaningful driver of renewal rates, then investments in provisioning quality have a return that extends beyond the direct cost saving on errors and failed visits. They have a revenue retention component. The installation that goes right first time is not just cheaper to deliver — it’s generating a customer relationship that starts from a better baseline and is more likely to continue.
This is a difficult number to quantify precisely, but it doesn’t need to be precise to be significant. If a 1% improvement in first-time installation success rate reduces churn rate by 0.2%, the revenue value of that improvement across the customer base is usually large enough to materially change the business case for provisioning investment.