A division manager posts a 22% ROI and refuses a project earning 18% because it would drag her ratio down. The same project clears the firm's 12% cost of capital. ROI just rejected value creation, and the exam wants you to name the fix.
Performance measures translate strategy into behavior. If the strategy is cost leadership, the measures track unit cost and yield. If it is differentiation, the measures track design cycle time and customer retention. Measures that ignore the strategy create motion without progress.
Three design principles govern every measure. First, link it to a strategic or operational goal. Second, feed it back fast enough to change behavior in the current period; a variance reported six months late teaches nothing. Third, tie it to the drivers of the element, not the element itself. To improve revenue, measure the revenue drivers (call volume, conversion rate, average order). To control cost, measure the cost drivers (machine hours, setups, batch size). Driver-level measures are leading; outcome measures are lagging.
Common mistakes
- Comparing ROIs across divisions with different inventory or depreciation policies. A LIFO division during inflation will out-ROI a FIFO division with identical operations because its invested capital is artificially low. Normalize the policies before ranking.
- Treating a negative segment margin as a drop signal. Allocated corporate overhead does not disappear when the segment closes; it reallocates to surviving segments. Drop only when controllable margin is negative or freed capacity has a better use.
- Using ROI to evaluate new projects. ROI rejects projects below the division's current ratio even when they clear the firm's hurdle. Use RI for project-acceptance decisions and reserve ROI for cross-segment comparison.
Bottom line
- ROI = Operating Income / Invested Capital, decomposing into return on sales × turnover; RI = Operating Income − (Required Rate × Invested Capital), expressed in dollars
- RI accepts any project above the hurdle; ROI rejects below-average winners, so a 21% division refuses an 18% project that still clears the firm's hurdle
- Controllable margin isolates costs the manager can influence and is the right number for manager evaluation
- Drop a product or customer only when controllable margin is negative, no strategic offset exists, and freed capacity has a better use
Exam shortcut
When a question pits ROI against RI on a new project, compute the project's standalone return and compare to the hurdle. If project return exceeds the hurdle, RI accepts. If project return is below divisional ROI but above the hurdle, ROI rejects and the answer is the ROI dysfunction. When a drop-the-customer question gives you allocated corporate overhead, strip it out and recompute on controllable margin before deciding.
The full lesson (about 2,562 words, 17 min read) adds 2 worked examples, all 6 common mistakes, a self-check, free in the app.
Learning objectives
- 1C3
Browse all free CMA Part 1 lessons or jump into free CMA Part 1 practice questions.