Gains on the three dimensions can offset each other; report them separately instead of merging into one score

What it is

Effectiveness, efficiency, and satisfaction can move in opposite directions: an efficiency gain bought with error rates, a satisfaction bump masking a completion drop. Merging the three into one total lets opposing movements cancel arithmetically into a false "no change overall." Standards and measurement practice therefore require component-wise reporting.

Why it happens

The offset lives in the arithmetic of weighted sums: any weighting choice silently asserts how much efficiency a unit of completion is worth—a value judgment that belongs to decision-making, not measurement. A composite score dresses those decision weights up as measurement results, and readers cannot recover the true direction and magnitude of components from the total. The dimensions also live on different event scales (errors discrete, time continuous, satisfaction ordinal), so direct merging adds dimensional noise on top.

Studying it

The reporting norm is three parallel columns: each dimension with its metric, baseline, change, and dispersion; where a combined judgment is needed, use explicit multi-criteria methods—weights published, sensitivity analyzed—rather than default equal averaging. In analysis, report cross-dimension correlations and the joint distribution of change directions—"efficiency up, errors up" is itself a diagnostically valuable pattern. At the meta level, counting how often published composites mask opposing changes is a methodological warning worth running.

Where it stops holding

Separate reporting does not forbid synthesis: budgeting and ranking decisions must synthesize—the requirement is only that synthesis be transparent: weights, method, and sensitivity inspectable. With many dimensions, component-wise reporting loses readability, and a layered structure (components plus a few key composites) serves better. Single composite scores do have a place where weights and purpose are explicitly agreed (an internal regression gate); the problem is presenting them as objective measurement output.

Applying it

• Fix a component-wise structure in report templates; any total score must ship with its weighting formula and raw component values.
• When decisions use a composite, declare the weights and their basis in the meeting, and run one sensitivity check on the weights.
• For summaries read by non-specialists, replace totals with component narratives so "flat" cannot mislead.

Related

• Same group: B5.09.1 Effectiveness is measured by accuracy and completeness of task outcomes, not by completion rate alone · B5.09.2 Efficiency's denominator is a resource—time, steps, or cognitive effort—and the report must say which · B5.09.3 Satisfaction is a subjective judgment, correlated with performance but not inferable from it
• Nearby: B5.02 Three Components · Q4 Measurement and Reliability
• Search terms: composite score · multi-criteria decision · reporting standards