Cost-to-serve

Two distributors each buy ₹10 lakh of shampoo a month at the same price. On paper they’re equally valuable. But:

• Distributor A orders once a month, a full truck of standard cases, collects on time and pays in 30 days.
• Distributor B orders 40 times a month in small mixed lots, wants special labels, rush deliveries to a remote area, and pays in 90 days.

B costs far more to serve: 40 order entries, 40 picks, 40 small deliveries, special packing, expediting and the cash tied up for longer. After those costs, A might earn a healthy margin while B barely breaks even, or loses money.

Cost-to-serve (CTS) makes this visible. It traces the costs of order handling, picking, packing, transport, special services, returns and payment terms to each customer, channel or order, usually with activity-based costing: count the activities each customer causes and cost them, instead of spreading logistics cost evenly by sales value. Time-driven ABC makes this practical by costing each activity at its time and rate.

The results often surprise people: some large customers turn out to be unprofitable, and some small, efficient ones turn out to be the best. The actions that follow are rarely “drop the customer”. More often they are:

• minimum order sizes, or charges for small or rush orders,
• fixed delivery days instead of on-demand drops,
• moving small customers to a distributor or an online channel,
• pricing or terms that reflect the real cost.

Time-driven activity-based costing

1. Cost of capacity: for each department, total cost per month divided by its practical capacity in minutes. A customer service team costing ₹6 lakh a month with 1,500 productive hours costs about ₹6.67 a minute.
2. Time per activity: how long each activity takes, with extra time for variations (an order with special labels takes 25 minutes instead of 10).
3. Count the activities each customer causes: orders, lines, deliveries, special requests, returns.
4. Multiply and add, together with directly traceable costs such as freight and the financing cost of payment terms.

What the numbers suggest

For B: a minimum order size or a small-order charge, fixed delivery days, a charge for special labels and rush deliveries, payment terms closer to A’s, or serving B through a distributor or an online ordering channel with lower handling costs. Each changes B’s costs or price to match the service it uses.

The whale curve

Plotting cumulative profit against customers ranked from most to least profitable often produces a whale curve: profit climbs above 100% of the total with the best customers, then falls back as the least profitable customers erode it. Kaplan and Narayanan (2001) describe how such curves show that a minority of customers generates more than all the profit while others destroy part of it. The shape points to where action pays: improving the tail, not just growing the head.

Why spreading costs by revenue misleads

Allocating logistics and service costs in proportion to sales (sometimes called “peanut-butter costing”, spread evenly) makes every customer look equally profitable per rupee of sales. It hides the cost of small orders, rush deliveries, special packaging, returns and long payment terms, and it subsidizes demanding customers at the expense of efficient ones. Activity-based costing replaces assumptions with drivers.

Service menus

Instead of one service for everyone, companies offer service menus: standard service at the base price, with extras (more frequent deliveries, rush orders, special labels, longer payment terms) priced separately or available only above certain volumes. Customers then choose the service they value enough to pay for, and demand for costly services falls to where it is worth it.

Strategic customers who lose money

Some unprofitable customers are worth keeping: a flagship retailer that builds the brand, a growing channel, a customer whose volume fills a plant. The analysis doesn’t decide for you; it shows the cost of the decision. The usual path is to reduce their cost to serve through collaboration (better forecasts, fuller orders, vendor-managed inventory) before changing terms.

Keeping it alive

A one-off study fades quickly. Building cost-to-serve into regular reporting by customer, channel and product (refreshed quarterly, linked to account plans and pricing reviews) keeps decisions grounded in current costs.