Customer profitability analysis uses cost-to-serve data to reveal true margin per client. The Whale Curve shows that 20% of customers generate 150-300% of profit while 10-20% destroy value. Revenue rankings mislead because high-revenue clients often consume disproportionate resources. TDABC models time consumed per transaction type to generate customer-level cost-to-serve without rebuilding management accounting. Segment clients into protect-and-grow, fix-or-renegotiate, grow-carefully, and review-and-exit categories based on true net margin.
The Whale Curve in Practice
In most companies, the relationship between customer rank and cumulative profitability follows what Kaplan and Cooper called the “whale curve”: the top 20% of customers generate 150–300% of total profit, the middle 60–70% roughly break even, and the bottom 10–20% destroy 50–200% of profit.
The aggregate P&L looks healthy. Underneath, the structure is not.
This is not a theory. It is a pattern we see consistently across manufacturing, logistics, professional services, and healthcare clients. The numbers vary by industry; the shape of the curve does not.
Why Revenue Misleads
A client generating €800K/year with complex bespoke requirements, frequent change requests, dedicated account management, and high return rates may be far less profitable than a client generating €300K/year with standardised orders, low service demand, and on-time payment.
Revenue-based ranking hides this. You invest more in the €800K client — more attention, more resources, more discounts to retain them — because they look important. In reality, you are subsidising them.
What a Proper Customer Profitability Analysis Looks Like
The starting point is cost-to-serve: the fully loaded cost of acquiring, onboarding, serving, invoicing, and retaining each customer. This includes:
- Direct costs (products/services delivered)
- Sales and account management time
- Customer service and support interactions
- Logistics and delivery complexity
- Credit risk and payment terms
- Returns, rework, and exception handling
Once you have cost-to-serve, the calculation is simple:
Client Net Margin = Revenue – COGS – Cost-to-Serve
The result typically ranks clients very differently from revenue alone.
The Segmentation That Follows
Once you have true margin per client, you can segment intelligently:
- Protect and grow: High margin, high revenue — your core
- Fix or renegotiate: High revenue, low/negative margin — often fixable with pricing or service model changes
- Grow carefully: Low revenue, high margin — these are your best growth opportunities
- Review and exit: Low revenue, low/negative margin — consuming resources with no return
The “review and exit” segment is often larger than expected. In our work with mid-market companies, it typically represents 15–25% of the customer base by count, and destroying 30–80% of the profit generated by the top tier.
The Data You Need (and Don’t Have)
The honest challenge: most companies don’t have cost-to-serve data at the customer level. They have revenue by customer. They may have COGS by product. They rarely have indirect cost allocation at the customer level.
This is exactly the problem TDABC solves. By modelling the time consumed by each transaction type and linking it to customers, you can generate cost-to-serve estimates without a full management accounting rebuild.
CostCTRL does this with SAF-T data and a structured cost pool setup that takes a few weeks to configure — not months.
A First Step You Can Take Today
Rank your top 20 customers by revenue. For each one, ask your team to estimate — not calculate, just estimate — how much service complexity they create relative to your average customer. Apply a simple multiplier to your overhead allocation.
You will immediately see 3–5 clients that look very different under that lens. That’s your signal to go deeper.
The Profitability Health Check assesses your current visibility into client-level profitability. 4 minutes. Immediate benchmark.