It's demonstrably cheaper to keep customers than churn them.
I have yet to see a demonstration as it applies to hosting in general and the small host in particular.
Even you then go on to say that you cannot say that.....
....you have to start making assumptions about customer decision drivers and the model becomes bogged down with subjectivity. i.e. did that customer leave because they found a better deal? Was the better deal including support or not... or just better support? Did support even matter? What did it cost me to support this customer incoming... then what outgoing? Was it the same "team" who provided onboarding support? What does reputation cost and what does potential damage to reputation for bad support cost? etc etc etc.
Your cost drivers are going to be all over the place and not really consistent between any two customers so you'd have to deal with them in aggregate then assume what the costs of a departing customer are and what factors you can link them to.
This would necessarly disqualify your intitial statement in addition to the ability to answer to my question, since the truthfulness of your statement does depend on the ability to answer the question.
But the gist of my pondering is one in the long run. Given that it is true -- that it costs more to acquire a new customer than to keep a current one -- and given the cost of keeping the current one is the cost of support, how do you demonstrate that?
If you do get the question answered, let me know because I have other burning questions that will probably share an answer. At top of mind,"how long is a piece of string"?
Try a ruler