Our scorecards keep growing — are we tracking too many KPIs, and holding people to numbers they can't actually move?

Our scorecards keep growing — are we tracking too many KPIs, and holding people to numbers they can't actually move?

Every incident adds a metric; no review ever removes one. The result is familiar: dashboards with dozens of KPIs per team, quarterly reviews that are mostly reporting theatre, and managers scored on outcomes — market share, attrition, an engagement index — that sit two or three causal steps beyond their reach. The live decision is not whether to measure (you must). It's a design choice with three options: keep accumulating (every number that can be tracked gets tracked and scored), cut to a disciplined accountability set (few, specific, controllable, with guardrails), or the pragmatic split most organisations never make explicit — measure many things, but judge people on few.

The damage from measurement overload is documented, recurring, and has a name-list. A rapid review and typology of performance measurement in healthcare (Li & Evans, 2022 — 41 peer-reviewed studies plus 147 practitioner interviews) catalogues the unintended consequences that show up wherever measurement regimes grow heavy: increased administrative burden, measure fixation and tunnel vision, misrepresentation and gaming, reduced morale, systemic dysfunction — up to decisions being driven by what is measured rather than by evidence. It echoes the classic public-sector list (Smith, 1995) of eight dysfunctions of published performance data: tunnel vision, suboptimization, myopia, measure fixation, misrepresentation, misinterpretation, gaming, and ossification. None of these require bad people — they are the predictable behaviour of reasonable people managed through too many consequential numbers.

The goal-setting science supports few targets — and warns about each one you add. As covered in our OKR brief, the strongest finding in this territory is that "challenging, specific, and concrete goals … are powerful motivators and boost performance … more than vague or abstract goals" (Höchli, Brügger & Messner, 2018, Frontiers in Psychology) — and the same literature is explicit that narrow "subordinate goals can lead to an overly narrow focus of attention," where people "run the risk of ignoring issues that are not specified by the goal." Multiply the metrics and you don't multiply the motivation; you fragment the attention. And every metric that carries consequences invites Goodhart's law — when a measure becomes a target, it ceases to be a good measure — one gaming surface per KPI.

The controllability research names the injustice your managers feel. A study of managers' opinions on the controllability principle (Giraud, Langevin & Mendoza, 2008, Management Accounting Research) found that when people are held accountable for results they do not control, they perceive their evaluation as unfair and become dissatisfied and demotivated. Follow-up work (Burkert, Fischer & Schäffer, 2011) adds a hierarchy effect: senior executives cope with uncontrollable factors far better than middle and junior managers — role clarity and resources buffer them — so uncontrollable KPIs do their worst damage exactly where most scorecards concentrate: the middle of the organisation.

The honest counterpoint: total insulation isn't the answer either. More recent accounting research ((Un)Controllability, Journal of Accounting Research, 2023) argues that some exposure to uncontrollable factors keeps employees attending to relevant information beyond their direct remit — pay a manager only on what they fully control and they stop looking at anything else. The lesson isn't "never reference uncontrollables"; it's that accountability and awareness are different instruments: judge people on the controllable few, keep the wider context visible without scoring it.

The real split isn't measure more vs. measure less. It's accountability vs. monitoring: a metric can be worth watching without being worth judging someone on. Most KPI bloat is that distinction, lost.

Split your measures, then cut. Keep a small accountability set — specific, challenging, controllable by the person scored on it, each with a guardrail. Keep a monitoring set that informs decisions but judges nobody. Delete what's left.

• What to rely on: the evidenced core of goal-setting — a few specific, ambitious goals per team; a controllability screen on anything wired to evaluation or pay; a guardrail metric beside every scored number ("how could someone hit this while making things worse?"); and human judgement at review time for the context no formula captures.
• What to avoid: cascading every dashboard number into personal scorecards; scoring managers on macro-outcomes they can't move (and watching your best middle managers quietly disengage); the one-incident-one-new-KPI ratchet; and measurement as its own department — if reporting effort rivals doing effort, the system is eating its purpose.
• The point that matters: nobody games a number they aren't judged on. Move a metric from the scorecard to the monitor and its Goodhart problem largely evaporates — which is why the accountability/monitoring split, not the total count on the dashboard, is the real design decision.

1. Run the two-question audit on every scored KPI. For each metric anyone is judged on: (a) can the person scored on it actually move its main driver? (b) what decision changed in the last two quarters because of it? A "no" on either sends it to monitoring or deletion.
2. Make the split explicit. Publish two lists per team: the accountability set (few — if everything is a priority, nothing is) and the monitoring set (watched in reviews, scored never). Say out loud that the second list carries no personal consequence — that sentence alone removes most of the gaming incentive.
3. Guardrail every survivor. For each accountability KPI, name the way it could be hit while making things worse, and pair it with the counter-measure (a quality, safety, or customer metric) — the same anti-gaming move as in our OKR brief.
4. Put controllability into the review conversation, not the formula. Windfalls and headwinds get discussed in calibration ("what was actually in your control this year?") rather than auto-scored — fair without teaching people to ignore the world outside their lane.
5. Institute a deletion rule. Any new KPI proposal must name the KPI it replaces or the review at which it will be re-justified. Scorecards only ever grow unless removal is somebody's explicit job.

The Gulf runs one of the world's most visible KPI cultures — which makes the design discipline more valuable here, not less.

The national context normalises big scorecards. As documented in our OKR brief, Vision 2030 is delivered through programs governed by defined objectives and KPIs with central, cadenced review (Vision Realization Programs) — so Gulf organisations are fluent in target machinery, and importing that fluency into a company usually means importing volume along with it. The discipline this brief argues for — few scored measures, the rest monitored — is how to keep the ambition without the dysfunction.

Where targets carry visibility, the failure modes bite harder. The gaming and misrepresentation risks in the typology above scale with the stakes attached to a number. In organisations where KPI reviews are high-ceremony and under-performance is high-consequence, expect exactly what the literature predicts: green dashboards, sandbagged targets, and energy flowing to what is counted rather than what counts.

Compliance metrics are a distinct category — treat them that way. Gulf organisations carry genuinely non-negotiable regulatory metrics (nationalisation ratios and similar). Those belong on the accountability list by law, not by choice. The design question is what sits alongside them: a mandated metric is not a licence for forty discretionary ones, and the controllability screen applies with full force to everything management adds on top.

Honest scope: we found no Gulf-specific studies of KPI overload or the controllability principle — the evidence above is Western (healthcare, public sector, corporate accounting research), and the regional read-across is informed inference from the documented mechanisms. The Vision 2030 KPI machinery is sourced; company-level conclusions from it are ours.

Library / open-licensed sources (Creative Commons; quoted from the publications themselves):

• Li, X. & Evans, J. M. (2022), Incentivizing performance in health care: a rapid review, typology and qualitative study of unintended consequences, BMC Health Services Research, 22:690 — original · licence: CC BY 4.0. 41 studies + 147 interviews; typology of unintended consequences: administrative burden, measure fixation, tunnel vision, misrepresentation and gaming, reduced morale, systemic dysfunction, decisions driven by measurement rather than evidence.
• Höchli, B., Brügger, A. & Messner, C. (2018), How Focusing on Superordinate Goals Motivates Broad, Long-Term Goal Pursuit: A Theoretical Perspective, Frontiers in Psychology — original · licence: CC BY. Specific, challenging goals outperform vague ones; narrow subordinate goals produce "an overly narrow focus of attention." (Shared with our OKR brief.)
• Höpfner, J. & Keith, N. (2021), Goal Missed, Self Hit: Goal-Setting, Goal-Failure, and Their Affective, Motivational, and Behavioral Consequences, Frontiers in Psychology — original · licence: CC BY. High, specific goals as an established performance tool — alongside the literature's turn to goal-setting's downsides. (Shared with the OKR brief.)

Cited findings (named and linked, not republished — these do not carry an open licence):

• Smith, P. (1995), On the unintended consequences of publishing performance data in the public sector, International Journal of Public Administration, 18(2–3) — the classic eight dysfunctions: tunnel vision, suboptimization, myopia, measure fixation, misrepresentation, misinterpretation, gaming, ossification. Cite-only.
• Giraud, F., Langevin, P. & Mendoza, C. (2008), Justice as a rationale for the controllability principle: A study of managers' opinions, Management Accounting Research, 19(1) — publisher. Managers held accountable for uncontrollables perceive evaluation as unfair; dissatisfaction and demotivation follow. Cite-only.
• Burkert, M., Fischer, F. M. & Schäffer, U. (2011), Application of the controllability principle and managerial performance: The role of role perceptions, Management Accounting Research, 22(3) — publisher. Senior managers withstand uncontrollable factors better than lower levels — the damage concentrates mid-organisation. Cite-only.
• The (Un)Controllability Principle: The Benefits of Holding Employees Accountable for Uncontrollable Factors (2023), Journal of Accounting Research — publisher. The counterpoint: some exposure to uncontrollables keeps attention on relevant context. Cite-only.
• Goodhart's law — "when a measure becomes a target, it ceases to be a good measure" (after economist Charles Goodhart). Background reference, shared with the OKR brief.

GCC context:

• Kingdom of Saudi Arabia — Vision 2030: Vision Realization Programs — the national objectives-and-KPIs delivery machinery. (Shared with the OKR brief.)
• Internal: Should we adopt OKRs? (the sibling brief — goal frameworks and gaming) · Should we monitor employee productivity? (measurement of people, and the watching-vs-judging line).