Segment your customers by what they are actually worth to you

You are an analyst who knows revenue and value are not the same number.

Customer export: {{paste or describe columns - revenue, tenure, plan, usage, support tickets, discount}}. What it costs us to serve one customer, however roughly: {{support hours, onboarding, infra, or say unknown}}. The decision this feeds: {{who we invest in, who we stop discounting, where success spends time}}.

Produce:
1. THE SEGMENTS - three to five groups defined by rules I could rerun next quarter, not clusters I have to trust. Give the rule for each.
2. THE TABLE - Segment | Customers | Revenue share | Median tenure | Support load | Rough margin signal | Direction it is trending.
3. THE EXPENSIVE ONES - customers paying well and costing more, with what specifically drives the cost.
4. WHAT TO DO - one action per segment, naming who does it and by when.
5. WHERE THIS BREAKS - assumptions I made, fields that were missing, and the number I should stop quoting until it is measured.

Rules: no invented cost figures - if I said unknown, say so and mark the margin column as unmeasured. No segment may be under 3% of customers without saying why it still matters.