Most SMEs discover their most 'profitable' customers are, on a fully-loaded basis, actually value-destroying — and their most 'demanding' customers are the ones the business quietly depends on. Cost-to-serve is where that inversion becomes visible.
Why is cost-to-serve a strategic decision, not a finance analysis?
The reversal
Cost-to-serve is usually delegated to finance and delivered as a spreadsheet. The higher-leverage frame is strategic: cost-to-serve decides which customers the business should double down on, which should be re-priced, which should be re-shaped, and which should be exited. Every one of those decisions is a strategy call disguised as an accounting analysis.
The insight stack
What actually moves the P&L
Fully-loaded, not just direct cost
Direct cost-to-serve understates by 30–70% in most SMEs — the difference is the operational overhead each customer actually consumes: sales cycle, onboarding, exception handling, escalations, executive attention. Load it all in, then look again at the segmentation.
Segmentation drives four strategic moves
The 2x2 of revenue-per-customer against cost-to-serve produces four quadrants: high-revenue, low-cost (invest and defend); high-revenue, high-cost (re-shape or re-price); low-revenue, low-cost (automate and retain); low-revenue, high-cost (exit or price to exit). Every customer has a strategic decision waiting.
Shaping beats exiting most of the time
The impulse in the high-cost quadrants is to fire the customer. Shaping — restructuring the engagement to eliminate the specific cost drivers — usually captures 60–80% of the value at a fraction of the political and revenue cost. Exit is a last resort, not a first move.
Case example
A £38M B2B services firm
A B2B services firm was carrying 340 customers on a blended gross margin of 41%. A fully-loaded cost-to-serve analysis showed the top 30 customers by revenue were operating at 62% margin, the middle 210 at 44%, and the bottom 100 at negative 8%. Rather than exit the bottom cohort en masse, the operator shaped the engagement: standardised onboarding, self-service reporting, restructured SLAs, and a repriced tier. Sixty-two customers accepted the reshape, eighteen left, twenty were re-priced upward. Twelve months later, the same customer base was generating 22% more gross profit with 24% less operational load — capacity that was redirected to acquiring high-margin segments.
Mini-playbook
Cost-to-serve rebuild
Fully load every category of operational cost against customers, not against products or channels.
Rebuild segmentation on revenue-per-customer × cost-to-serve.
For each quadrant, name a strategic move: invest, reshape, automate, or exit.
Shape before you exit — most high-cost customers can be restructured, not fired.
Redeploy the freed capacity into the invest-and-defend quadrant; do not re-absorb it silently.
How Strategy Labs installs this
Anchored to Cost-to-serve optimisation
CAE runs cost-to-serve as one of the four operating artefacts, integrating the analysis with the operating model, performance management, and process re-engineering workstreams rather than delivering it as a standalone finance study.
PDC hosts the fully-loaded cost model, quadrant maps, and customer-shaping playbooks so the strategic moves accumulate into a repeatable operating discipline rather than a one-off review.
Frequently asked
Related questions executives ask
- How often should cost-to-serve be refreshed?
- Annually at customer-level, quarterly at segment-level. More frequent customer-level refresh usually gets ignored; less frequent segment-level refresh lets the shape of the book drift undetected.
- How do we avoid double-counting operational overhead?
- Allocate through a single, published driver per cost category — support hours, incident count, sales-cycle duration — and reconcile total allocated cost against the P&L before drawing conclusions. Multiple ad-hoc allocations are where credibility disappears.
- Should we exit unprofitable customers immediately?
- Almost never. Shape first, re-price second, exit third. Immediate exits leak revenue faster than the operational savings can catch up, and burn goodwill in referral networks that were often the acquisition channel in the first place.
Over to you
If you knew, precisely, which fifth of your customer base was destroying operating value — would your growth plan for next year still look the same?
Continue reading
More Operational Strategy briefings
What is an operating strategy — and why do most companies not have one?
Most companies have a strategy deck. Very few have an operating strategy — a document that translates the strategy into the specific decisions the operating team makes every week. The gap is where strategy dies.
Read briefingHow do you design a revenue operating model that produces acquisition, retention, and expansion as one compounding system?
Great commercial teams produce spikes. Great revenue operating models produce compounding. The difference is that the model treats acquisition, retention, and expansion as one system — designed by the business, not defended by the sales leader.
Read briefingWhat commercial enablement scaffolding does a growing business actually need — and what is the difference between enablement and tooling?
Commercial teams stop being the bottleneck when the system carries the routine work. Enablement is the scaffolding — not the tooling — that makes it possible. The tooling shows up in every RFP. The scaffolding shows up in the P&L.
Read briefing
Discussion
(…)Comments are moderated before appearing. Your email is only used for moderation and is never shown publicly.
Loading discussion…