TL;DR
Cost to Serve is the total operational cost of serving a single customer over a defined period — including support, hosting, fulfilment, payment processing, and CS allocations. For B2B SaaS, healthy annual cost to serve is 8–18% of customer ARR; for D2C, healthy cost to serve per order is 6–14% of order value. Cost to Serve is the operational complement to CAC and a key input to <a href="/glossary/contribution-margin" class="text-brand-600 underline decoration-brand-200 underline-offset-2 hover:text-brand-700">contribution margin</a> per customer.
What is cost to serve?
Cost to Serve is the total operational cost a business incurs to serve a single customer over a defined period — typically reported annually for B2B SaaS and per-order for D2C. It includes customer support, infrastructure costs (hosting, bandwidth), payment processing, fulfilment, and allocated customer-success headcount.
It is the operational complement to CAC: where CAC measures the cost of bringing a customer in, cost to serve measures the ongoing cost of keeping them. Together they determine contribution margin per customer — the unit-economics number that ultimately drives profitability.
How to calculate it
For B2B SaaS (annual): Cost to Serve = (Customer Support + Hosting/Infra + Payment Processing + CS Headcount Allocation + Account Mgmt Allocation) / Active Customers For D2C (per order): Cost to Serve = (Pick + Pack + Ship Cost + Customer Service Allocation + Payment Processing + Returns Processing) / Orders Hosting/infra allocation typically uses customer-tier-weighted apportionment. CS headcount allocation uses tickets-per-customer or revenue-weighted apportionment.
Benchmarks
| Category | Healthy CTS | Caution |
|---|---|---|
| B2B SaaS — SMB | 12–18% of ARR | >22% |
| B2B SaaS — Mid-market | 10–15% of ARR | >18% |
| B2B SaaS — Enterprise | 8–12% of ARR | >15% |
| D2C consumables | 8–14% of AOV | >18% |
| D2C apparel (with returns) | 12–20% of AOV | >25% |
Common pitfalls
- 1. Allocating support costs uniformly across customers. A small fraction of customers typically generates a majority of support load. Uniform allocation hides which customer segments are unprofitable. Allocate by ticket volume or weighted by tier.
- 2. Excluding payment processing. 2.5–3.5% of revenue across all customers — not negligible at scale. Always include in cost-to-serve calculations.
- 3. Treating cost to serve as fixed. CTS scales differently from revenue: support costs scale with customer count, hosting scales with usage, fulfilment scales with order count. Track each component separately to see scaling dynamics.
Why it matters
Cost to Serve is often the missing variable in unit-economics conversations. Brands optimise CAC and chase LTV but rarely instrument cost to serve at the customer-segment level — meaning they don't actually know which segments are profitable.
Healthy operations track contribution margin per customer = LTV − CAC − cumulative cost to serve − cost to retain. Cost to serve is the lever most operators have the most slack on; it's also the lever least often examined.
Related concepts
CAC is the acquisition-side complement. Cost to Retain is the retention-spend complement. Contribution margin is the per-customer outcome that combines all three. Cost to Serve directly affects gross margin for service-heavy businesses.
At a glance
- Category
- Profit Intelligence
- Related
- 5 terms
Frequently asked questions
What's a healthy cost to serve?
B2B SaaS SMB: 12–18% of ARR. Mid-market: 10–15%. Enterprise: 8–12%. D2C consumables: 8–14% of AOV. D2C apparel (with returns): 12–20%. Healthy ranges depend heavily on category and motion — benchmark within both.
Should you allocate hosting per-customer?
Yes — but use weighted apportionment, not uniform. Customer-tier-based apportionment (enterprise customers consume more) or usage-based apportionment (heavy users consume more) is more accurate. Uniform allocation hides cross-subsidisation between tiers.
How do you reduce cost to serve?
Three main levers: (1) self-serve / deflection — knowledge base, in-product guidance, AI support; (2) cohort segmentation — identify which segments drive disproportionate cost; (3) tier-pricing alignment — ensure pricing reflects service load. Most operators have meaningful headroom on all three.
Sources
- B2B SaaS operations benchmarks
- Shopify D2C operating data
- Fairview customer data (2025)
Fairview is an operating intelligence platform that computes cost to serve at customer-segment level — joining support ticket volume, hosting allocation, payment processing, and fulfilment cost — so unit-economics conversations include the operational cost most operators leave out. Start your free trial →
Siddharth Gangal is the founder of Fairview. He built the segment-level cost-to-serve layer after watching a B2B SaaS company optimise heavily on LTV:CAC for two years before discovering that 35% of their SMB customers were contribution-margin-negative on a fully-loaded basis — the cost-to-serve dynamics had completely changed the unit economics, but nobody was measuring it.
See it in Fairview
Track Cost to Serve automatically.
14-day free trial. No credit card. First data source connected in 5 minutes.