Management Accounting for Service Pricing: Activity-Based Costing

Why does pricing with a fixed overhead percentage mislead?

Many service firms price with a simple formula: the contract’s direct cost, plus a fixed percentage to cover overheads, plus a profit margin. The problem is that the percentage assumes every customer uses administrative support, coordination and reporting to the same degree, which rarely happens.

Traditional costing often spreads overheads using a single factor, such as labour hours, that has no causal link to the work performed. ^[1] As ACCA’s example shows, the result is that low-volume items with heavy set-up are undercosted, while high-volume, repetitive ones are overcosted. ^[4] In other words, the easy customer quietly subsidises the expensive one.

What is activity-based costing?

It is a method that traces direct and indirect costs to the services, channels and customers that cause them. ^[1] It works in two stages: resources such as salaries, rent and systems are assigned to the activities that consume them, and activity costs are then assigned to services and customers according to their use. ^[1]

IMA’s statement distinguishes three types of driver: ^[1]

• Resource driver: measures how much of a resource an activity consumes, such as staff time.
• Activity driver: measures an activity’s output, such as the number of orders processed.
• Cost object driver: measures one cost object’s consumption of another, such as a customer buying a mix of services.

The governing principle is that cost should vary directly with the quantity of the driver, and the driver must be measurable. ^[1] ACCA summarises the calculation in four steps: split overheads into activity cost pools, identify each pool’s driver, calculate a cost per unit of driver, and assign costs according to use. ^[4]

When is it worth the effort?

The method takes time and effort to establish what causes costs, ^[4] so it pays off most when these conditions apply:

• Overheads are material relative to total cost. ^[4]
• Services or customers are diverse, and some demand customisation or special handling. ^[4]
• You suspect some customers or contracts are unprofitable despite their size.
• The decision is clear: pricing, a contract review or a process improvement.

Although the method began in manufacturing, CIMA notes that its ideas apply to service firms and government bodies, ^[5] which makes it useful for public entities that need to understand the cost of their services before reviewing fees.

The cost hierarchy: what should be assigned to a service?

Not every overhead should be assigned. IMA’s statement separates transaction or batch costs, product- or customer-sustaining costs, and the costs of sustaining the business as a whole. The last group, such as the accounting close and regulatory filings, is not assigned to services or customers because there is no causal link. ^[1]

The same goes for unused capacity: the statement recommends tracing it to its own separate line rather than spreading it over the remaining services. ^[1] This matters for pricing, because loading it onto customers raises your prices for capacity nobody is using.

The time-driven version: a lighter method for service firms

Conventional ABC ran into a practical problem. It relied on interviews and surveys to split staff time across activities, which were costly, irritating for employees and subjective, and re-estimating the model was expensive, so many companies abandoned it or rarely updated it. ^[3]

In 2004 Kaplan and Anderson therefore proposed the time-driven version, which needs only two estimates: ^{[2, 3]}

1. Cost per unit of capacity: the department’s cost divided by the minutes its staff actually have available.
2. Transaction time: how many minutes each transaction or activity takes.

It handles variety through time equations that add time when a particular feature is present, such as a rush or customised order. ^[3] It also shows unused capacity explicitly instead of hiding it. ^[2]

Hypothetical example: pricing two maintenance contracts

A facilities maintenance company’s coordination and customer service department costs AED 450,000 a quarter, which it currently spreads evenly at AED 75 per service request. The department has 10 staff, and the manager estimates that each has 25,000 minutes a quarter actually available after breaks, training and meetings.

• Available capacity: 10 × 25,000 = 250,000 minutes.
• Cost per minute: 450,000 ÷ 250,000 = AED 1.80.
• Time equation: 20 minutes per request, plus 25 minutes for an urgent request, 60 minutes for a standard monthly report and 180 minutes for a bespoke weekly report.

Customer A is charged almost twice what it consumes, so the company may lose it to a cheaper competitor. Customer B is charged below cost, and its contract may look profitable when it is not. Part of the gap between AED 75 and the new figures is unused capacity cost, which now appears as a separate line.

Steps for building a first model

IMA’s statement suggests starting from the decisions the system will support and working backwards, and choosing between a pilot and a full implementation built through iterative prototypes. ^[1] The project needs a cross-functional team and training for data providers, because the results shift costs between departments and may meet resistance. ^[1]

For a practical start:

1. Pick one high-cost support department, such as coordination, billing or customer service.
2. Take its cost from the general ledger and estimate its actually available minutes.
3. Identify 5 to 10 main activities; materiality, not completeness, is the test. ^[1]
4. Estimate each activity’s time by observation or from system logs, and add the features that change the time.
5. Extract transaction volumes per customer from the ordering or billing system.
6. Show the results to operations managers and ask: does this match your experience?

Remember that approximate accuracy is enough; perfect precision is not required. ^[1]

From cost to price

The model does not set the price. IMA’s statement describes pricing as a management decision that is typically market based; the model’s job is to reveal the true margin and improve the estimates behind quotations. ^[1] Once the results are in, management’s options include:

• Re-pricing costly activities or adding surcharges for them, such as urgent requests or bespoke reports. ^[1]
• Improving the process itself to cut activity time rather than raising the price.
• Revising contract terms at renewal so that extra services are priced as separate items.

Beware of dropping an unprofitable customer without reducing the resources linked to it. Dropping it only frees capacity while the resource cost remains, so it is spread over those who remain and starts what IMA calls an overhead death spiral. ^[1]

Conclusion

Activity-based costing answers one question before pricing: what does this customer or service actually cost us? Start with one support department and a simple time-driven model, separate unused capacity, then use the results to redesign offers and contracts, not just to raise prices.