FinanceIntermediate8 min read

Activity-Based Costing

Activity-Based Costing (ABC) assigns indirect costs to products, customers, channels, or geographies based on the actual activities they consume — not based on broad allocation bases like revenue or labor hours. ABC was formalized by Robert Kaplan and Robin Cooper at Harvard Business School in the late 1980s as a response to traditional cost accounting, which systematically over-costed high-volume simple products and under-costed low-volume complex products. ABC asks: 'What activities does this product/customer/order actually trigger, and what does each activity cost?' The output is shockingly different unit economics from what your P&L shows. Companies that run ABC routinely discover that 20-40% of their products or customers are unprofitable on a fully-loaded basis — invisible to traditional GAAP costing.

Also known asABCActivity-Based Cost AccountingDriver-Based CostingCost Driver Analysis

Challenge a friend Browse library

The Trap

The trap is over-engineering ABC into a 200-activity, 50-driver behemoth that takes a year to build and is obsolete the day it ships. Kaplan himself published a 2004 update — Time-Driven ABC — explicitly because most ABC implementations collapsed under their own complexity. The other trap: running ABC and then doing nothing with the findings. ABC routinely surfaces uncomfortable truths (your biggest customer is unprofitable; your hero product is a loss leader). Companies that build ABC dashboards but lack the political will to act on them have wasted the investment.

What to Do

Run ABC at the right altitude: aim for 8-15 activities, 3-5 drivers, and a quarterly refresh — not a continuous real-time system. Pick the 80/20 cost categories that traditional accounting allocates badly: customer service, logistics, R&D, sales support, complexity-driven manufacturing overhead. For each, identify the driver (e.g., service tickets per customer, SKUs per channel, line items per order). Then re-allocate. Then ACT: kill or reprice the unprofitable products/customers ABC reveals. ABC without action is just an interesting report.

Formula

ABC Unit Cost = Σ (Activity Cost ÷ Activity Driver Volume) × Driver Consumption per Unit

In Practice

Robert Kaplan's original ABC research at HBS (1988-1992) studied dozens of manufacturers and consistently found the same pattern: traditional cost accounting under-costed low-volume specialty products by 20-50% and over-costed high-volume commodity products by similar amounts. One famous case study (Schrader Bellows) found that of 7,500 SKUs, the bottom 30% were profit-destroying on an ABC basis but appeared profitable under the existing standard cost system. After implementing ABC and pruning, the company increased operating profit by 40% with lower revenue. ABC routinely produces this 'less revenue, more profit' outcome — which is why it threatens managers whose comp is tied to revenue.

Pro Tips

01
Robert Kaplan's Time-Driven ABC (TDABC, 2004) is a major simplification: instead of surveying employees on % time per activity, you estimate the time each transaction takes and multiply by the cost per minute of capacity. Far easier to maintain. Most modern ABC implementations should default to TDABC unless there's a specific reason not to.
02
The single most surprising finding ABC consistently produces: small customers are usually MORE profitable per dollar than large customers, because large customers extract disproportionate service, customization, and discount. This contradicts every salesperson's intuition. Run the numbers anyway.
03
Build ABC for decisions, not for reports. If the only output is a dashboard, the project will fail. Tie ABC outputs directly to specific decisions: which customers to fire, which SKUs to discontinue, which channels to deprioritize. Decisions create the political cover to maintain the model.

Myth vs Reality

Myth

“ABC requires a custom IT system and a dedicated team”

Reality

Time-Driven ABC can be built in Excel for most companies under $500M revenue. The bottleneck is usually clarity on activities and drivers, not data infrastructure. The 'we need a system first' argument is often a deflection from the harder work of agreeing on activity definitions.

Myth

“Our standard cost accounting is good enough”

Reality

Standard costing was designed for a world of high-volume manufacturing of similar products. In any business with product mix complexity, customer customization, or service intensity (which is most modern businesses), standard costing is systematically wrong by 15-40% per unit. The error is invisible — until you lose money on customers you thought were your best.

Try it

Run the numbers.

Pressure-test the concept against your own knowledge — answer the challenge or try the live scenario.

🧪

Knowledge Check

A company allocates $10M of customer service overhead to 1,000 customers based on revenue share. Customer A generates 10% of revenue but consumes 25% of service tickets. Customer B generates 5% of revenue but consumes 1% of tickets. Under ABC, what's the directional reallocation?

Industry benchmarks

Is your number good?

Calibrate against real-world tiers. Use these ranges as targets — not absolutes.

ABC Reveals Unprofitable Customers (% of customer base)

Cross-industry ABC implementations — typical findings in first audit

Healthy Portfolio

< 10%

Average

10-20%

Concerning

20-30%

Profit Trap

30-40%

Critical Restructure Needed

> 40%

Source: Kaplan & Cooper, Cost & Effect (HBS Press, 1998)

Real-world cases

Companies that lived this.

Verified narratives with the numbers that prove (or break) the concept.

🔧

Schrader Bellows (Kaplan ABC research)

Late 1980s

success

Schrader Bellows manufactured 7,500 valve and pneumatic SKUs and used standard costing. Robert Kaplan's HBS research team implemented ABC and discovered that ~30% of SKUs were destroying value once true overhead consumption (setups, complexity, inventory carrying) was assigned. Standard costing had hidden the losses by spreading overhead via direct labor hours — a denominator that was barely related to actual overhead drivers. After ABC-driven SKU rationalization, the company shed unprofitable lines and increased operating profit despite lower revenue.

SKUs Analyzed

7,500

% Unprofitable Under ABC

~30%

Operating Profit Lift Post-Pruning

+40%

Revenue Change

Lower (intentional)

Standard costing systematically lies in companies with product mix complexity. ABC reveals the truth and enables the painful but profitable decision: shrink revenue to grow profit. This case shaped Kaplan and Cooper's foundational ABC framework.

Source ↗

📦

Hypothetical: Mid-Cap Distributor

2021-2023

success

A $300M industrial distributor implemented Time-Driven ABC across customer service, warehouse, and logistics functions. The audit revealed that the top 3 customers (40% of revenue) consumed 65% of variable overhead due to special pricing terms, custom packaging, and high return rates. After repricing two of the three customers, one walked. Revenue dropped 12%, but EBITDA grew 22% because the freed capacity served more profitable mid-tier accounts.

Top-3 Customer Revenue Share

40%

Top-3 Customer Overhead Share

65%

Revenue Change Post-Action

-12%

EBITDA Change Post-Action

+22%

ABC consistently reveals that the biggest customer is rarely the most profitable customer. The courage to act on the finding — repricing or firing the customer — is where most companies fail. The data is easy; the politics are hard.

Decision scenario

The Big Customer Audit

You're VP Finance at a $200M B2B services firm. ABC analysis just landed on your desk: your largest customer (22% of revenue, $44M) is consuming 38% of cost-to-serve due to bespoke SLAs, weekly executive reviews, and custom reporting. Their fully-loaded margin is -3%. Their contract renewal is in 90 days.

Customer Revenue

$44M (22% of total)

Cost-to-Serve Share

38%

Reported Margin (allocated)

+18%

ABC True Margin

-3%

Renewal Window

90 days

Decision 1

Your CRO says firing the customer would crater the year and cost the sales team their bonuses. Your CEO says he trusts the ABC data but is 'not ready to lose 22% of revenue overnight.' You have three options.

Renew at the same price — trust the legacy P&L, the loss is theoreticalReveal

You renew quietly. The customer extracts another 5% in implicit concessions over the next 18 months. By year 3, ABC margin is -8%. The customer eventually churns to a competitor anyway and your company has spent 3 years subsidizing their business at the cost of investment in profitable accounts. Total opportunity cost: ~$15M.

3-Year Cumulative Loss: ~$15MCustomer Outcome: Churned to competitor

Negotiate a 15% price increase + remove three custom SLAs that drive 40% of the cost-to-serve gap. Be transparent that the current pricing is unsustainable.Reveal

Tense 60-day negotiation. The customer pushes back hard initially but agrees to a 12% increase + dropping two of three custom SLAs. ABC margin moves from -3% to +9%. Customer relationship is briefly strained but professionally intact. You've recovered ~$5M of annualized economic loss without losing the revenue. The sales team gets paid a partial bonus. Best of three options.

ABC Margin: -3% → +9%Annualized Profit Recovery: +$5MCustomer Retained: Yes, terms restructured

Decline to renew — the customer is unprofitable and will only get worse. Free the capacity for higher-margin customers.Reveal

Bold but premature. You lose $44M of revenue overnight. Replacing it takes 14 months. Sales morale collapses. Your CEO loses confidence in ABC because the action felt too extreme. You eventually replace the revenue with higher-margin customers, but the transition pain was avoidable. The right answer wasn't binary keep-or-fire — it was renegotiate.

Revenue Impact (Year 1): -$44MReplacement Time: 14 monthsCEO Confidence in ABC: Damaged

Related concepts