A quarterly diagnostic for detecting where metrics have become targets that corrupt their own purpose, and installing the counter-gaming patterns that keep them honest
Published 19 March 2026.
What the Map Does
The Goodhart map is a quarterly diagnostic tool that identifies which of the team's metrics are at risk of becoming targets that corrupt their own purpose, and installs counter-gaming patterns to keep them honest. It is designed to run in a two-hour team meeting — light enough to sustain quarterly, rigorous enough to catch moral hazard before it metastasizes into capture.
The map works in four steps: identify high-stakes metrics, map the gaming pathways that threaten them, install counter-gaming patterns to harden the structure, and rotate metrics to prevent any single measure from becoming sacred.
Step 1: Identify High-Stakes Metrics (20 minutes)
List every metric the team uses to make decisions about people, resources, or outcomes. For each, ask: Does hitting this metric affect compensation, promotion, evaluation, or resource allocation? If yes, it is high-stakes. A metric is hazardous when people are evaluated based on it, when optimizing for it without serving the underlying value is possible, and when gaming is hard to detect.
In a typical team, three to five metrics meet these criteria.
Table 2.3B.3-1 — High-stakes metric identification
| Metric | What It Measures | Stakes | Who Benefits |
|---|---|---|---|
| [Metric name] | [What it tracks] | [Compensation, evaluation, promotion] | [Roles affected] |
Step 2: Map the Gaming Pathways (40 minutes)
For each high-stakes metric, ask: How could someone satisfy this metric while failing the underlying value it was meant to track?
Table 2.3B.3-2 — Gaming pathway analysis
| Field | Content |
|---|---|
| The metric | [What is being measured] |
| The value | [What underlying purpose this metric was designed to serve] |
| The gaming pathway | [How someone could hit the metric without serving the value] |
| Gaming visibility | [Easy to hide, or leaves visible traces?] |
| Structural exposure | [Which counter-gaming patterns would defend against this pathway] |
Worked example — velocity: The metric is story points delivered per sprint. The value is reliable, maintainable capacity. The gaming pathway includes inflating estimates, marking work done before integration, and shifting effort from unmeasured dimensions (documentation, technical debt) to measured ones (shipping features). Structural exposure: randomized audits (spot-check five story points mid-project), separation of duties (different team validates completion).
Step 3: Install Counter-Gaming Patterns (40 minutes)
For each gaming pathway, select one or more of the six patterns:
Table 2.3B.3-3 — Counter-gaming pattern selection
| Pattern | How It Works | When to Use |
|---|---|---|
| Randomized audits | Audit at unpredictable times and in unpredictable ways | When gaming depends on predictability |
| Adversarial testing | External or independent team tests whether the metric still tracks the underlying value | When metric may have decoupled from purpose |
| Capped leverage | No single metric controls more than 40% of compensation or evaluation | For all high-stakes metrics |
| Separation of duties | People who benefit from a metric do not set it, measure it, or decide what satisfies it | For compensation and evaluation metrics |
| Disclosure rules | Publish how metrics are calculated, what assumptions are made, when definitions change | For all high-stakes metrics |
| Compensation clawbacks | Tie payoff to long-horizon outcomes; bonuses not fully paid until stability confirmed | For metrics with strong short-term gaming incentives |
For each high-stakes metric, document the selected pattern and implementation detail.
Step 4: Schedule Review and Rotation (5 minutes)
Schedule the next Goodhart map for ninety days. At the next cycle: are these still the right metrics? Have gaming pathways changed? Are counter-gaming patterns holding?
Plan metric rotation: if a metric has been stable for two quarters, rotate it out of focus and bring a different high-stakes metric in. This prevents metrics from becoming sacred and ensures all high-stakes metrics are subject to scrutiny.
Log the results in the team's private artifacts layer: metrics identified, gaming pathways found, patterns installed, and what changed from last quarter.
Change Log
- 19 Mar 26 — Initial publication.