AbstractResearch SummaryWe discuss two research design considerations that jointly influence the choice of financial performance metrics in strategy research: (a) expected temporal payoff of the strategic choice and (b) source of variation invoked in the research design (i.e., within‐firm vs. between‐firm comparisons). We map existing performance metrics commonly used in the research literature to these considerations, and highlight the lack of performance metrics well suited for the combination of strategies with not‐well‐defined temporal payoffs and within‐firm research designs. To remedy this, we introduce a value partitioning methodology that provides a performance metric we call dynamic value. We apply this methodology and demonstrate how it generates additional insights about how cash holdings affect firm performance.Managerial SummaryThis article introduces the value partitioning methodology for interpreting two distinct elements of stock market valuations: (a) the firm's implied future value from continuing operations in a “steady state” and (b) the market's estimate of dynamic future value associated with the firm's strategic choices or competitive positioning. We describe situations where the latter element is useful for measuring strategy performance. We apply this methodology and demonstrate how it generates additional insights about how cash holdings affect firm performance.