Abstract
This paper studies a manager's attempt to maximize his firm's discounted expected profits by choosing what strategic actions to select and what performance measurement system to employ in a setting where the manager is uncertain about what variables "drive" the firm's profits, the firm's profit drivers remain stationary over time, and strategic actions differ in the amount of information they produce about the firm's profit drivers. For each available performance measurement system, this paper identifies necessary and sufficient conditions for experimentation-that is, deviating from the firm's short-run expected profit-maximizing action - to be optimal. In addition, the paper determines what factors influence a firm's preferred performance measurement system, and it explains why the preferred performance measurement system is likely to change over time.
Original language | English (US) |
---|---|
Pages (from-to) | 1624-1637 |
Number of pages | 14 |
Journal | Management Science |
Volume | 50 |
Issue number | 12 |
DOIs | |
State | Published - Dec 1 2004 |
Keywords
- Bandit problems
- Business models
- Experimentation
- Learning-by-doing
- Profit drivers
ASJC Scopus subject areas
- Strategy and Management
- Management Science and Operations Research