Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply organizational ecology (Hannan and Freeman) to analyze population-level dynamics of organizational founding, failure, and selection. Use this skill when the user needs to explain industry-level patterns of birth and death rates, analyze structural inertia and liabilities of newness or aging, evaluate why adaptation is rare relative to selection, or when they ask 'why do most startups fail', 'why is this industry dominated by old firms', or 'why do organizations resist change despite enviro
.claude/skills/asgard-ai-platform-grad-org-ecology/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -22% | 0% |
Organizational ecology applies population-level evolutionary logic to organizations. Rather than asking why individual firms succeed or fail, it asks why certain organizational forms proliferate or decline across populations. The theory emphasizes selection over adaptation: environmental change is met more by the founding of new organizations and the failure of old ones than by incumbent transformation.
IRON LAW: Selection operates on POPULATIONS, not individual organizations —
organizational change is driven more by founding and failure than by
adaptation. Any analysis that assumes incumbent firms can readily
transform themselves violates the structural inertia thesis.Key assumptions:
Identify the organizational form and population boundaries. Populations share a common form (technology, structure, market orientation).
| Phase | Density | Legitimation | Competition | Net Effect | |-------|---------|-------------|-------------|------------| | Early | Low | Rising fast | Low | Founding rate increases | | Growth | Medium | High | Rising | Peak founding, rising failure | | Mature | High | Saturated | Intense | Founding slows, failure rises | | Decline | Falling | Declining | Easing | Population contracts |
| Factor | Description | Impact | |--------|-------------|--------| | Structural inertia | Internal (sunk costs, politics, norms) and external (barriers, legitimacy) pressures resist change | Limits adaptation | | Liability of newness | New organizations lack routines, legitimacy, and stable relationships | Higher early failure rate | | Liability of smallness | Small organizations have fewer resources to buffer environmental shocks | Size-dependent mortality | | Liability of adolescence | Organizations fail after initial resources deplete but before routines establish | Delayed mortality peak | | Liability of aging | Older organizations accumulate structural rigidity | Vulnerability to environmental shifts |
Analyze founding rates, failure rates, and the relative contribution of selection vs. adaptation to population-level change.
markdown## Organizational Ecology Analysis: [Context] ### Population Definition - Organizational form: [description] - Population boundaries: [geographic, temporal, industry] - Current density: [approximate number of organizations] ### Density Dependence Assessment - Current phase: [early / growth / mature / decline] - Legitimation level: [H/M/L] - Competition intensity: [H/M/L] - Predicted trajectory: [founding/failure rate trends] ### Inertia and Liability Assessment | Factor | Severity | Evidence | |--------|----------|----------| | Structural inertia | [H/M/L] | [specific evidence] | | Liability of newness | [H/M/L] | [specific evidence] | | Liability of smallness | [H/M/L] | [specific evidence] | | Liability of adolescence | [H/M/L] | [specific evidence] | | Liability of aging | [H/M/L] | [specific evidence] | ### Selection vs. Adaptation - Proportion of change via selection (founding + failure): ... - Proportion of change via adaptation (incumbent transformation): ... ### Implications 1. [What the population-level dynamics predict for this industry] 2. [Whether new entrants or incumbents are favored by current conditions]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 48,157 | 35,130 | -27% | 1 | 1 | 0% | 7,739 | 4,531 | -41% | 0 | 0 | — |
case-02 | fail→pass | 30,368 | 20,520 | -32% | 1 | 1 | 0% | 4,483 | 4,496 | +0% | 0 | 0 | — |
case-03 | fail→pass | 26,207 | 17,007 | -35% | 1 | 1 | 0% | 4,039 | 3,707 | -8% | 0 | 0 | — |
case-04 | fail→fail | 25,317 | 17,300 | -32% | 1 | 1 | 0% | 4,045 | 4,080 | +1% | 0 | 0 | — |
case-16 | pass→pass | 47,940 | 16,276 | -66% | 1 | 1 | 0% | 6,654 | 3,452 | -48% | 0 | 0 | — |
case-05 | fail→fail | 20,267 | 22,346 | +10% | 1 | 1 | 0% | 2,963 | 4,725 | +59% | 0 | 0 | — |
case-06 | fail→fail | 51,038 | 23,679 | -54% | 1 | 1 | 0% | 8,226 | 5,350 | -35% | 0 | 0 | — |
case-07 | pass→pass | 45,446 | 19,430 | -57% | 1 | 1 | 0% | 6,920 | 4,393 | -37% | 0 | 0 | — |
case-08 | fail→pass | 29,317 | 14,275 | -51% | 1 | 1 | 0% | 4,358 | 3,430 | -21% | 0 | 0 | — |
case-09 | fail→pass | 36,546 | 15,689 | -57% | 1 | 1 | 0% | 5,557 | 3,506 | -37% | 0 | 0 | — |
case-10 | fail→fail | 43,361 | 19,453 | -55% | 1 | 1 | 0% | 6,268 | 4,282 | -32% | 0 | 0 | — |
case-11 | fail→fail | 40,143 | 22,265 | -45% | 1 | 1 | 0% | 6,254 | 4,446 | -29% | 0 | 0 | — |
case-12 | pass→pass | 26,210 | 18,011 | -31% | 1 | 1 | 0% | 3,822 | 3,792 | -1% | 0 | 0 | — |
case-13 | fail→pass | 42,787 | 24,717 | -42% | 1 | 1 | 0% | 5,882 | 4,603 | -22% | 0 | 0 | — |
case-14 | fail→pass | 44,749 | 17,534 | -61% | 1 | 1 | 0% | 6,123 | 3,572 | -42% | 0 | 0 | — |
case-15 | fail→pass | 38,342 | 13,858 | -64% | 1 | 1 | 0% | 6,387 | 3,465 | -46% | 0 | 0 | — |
case-17 | fail→pass | 33,618 | 17,986 | -46% | 1 | 1 | 0% | 4,829 | 3,556 | -26% | 0 | 0 | — |
case-18 | fail→pass | 38,476 | 23,361 | -39% | 1 | 1 | 0% | 5,305 | 4,325 | -18% | 0 | 0 | — |
case-19 | pass→pass | 28,716 | 16,780 | -42% | 1 | 1 | 0% | 4,129 | 3,608 | -13% | 0 | 0 | — |
case-20 | fail→pass | 24,681 | 17,261 | -30% | 1 | 1 | 0% | 3,809 | 3,707 | -3% | 0 | 0 | — |
case-21 | pass→pass | 23,292 | 17,097 | -27% | 1 | 1 | 0% | 3,357 | 3,874 | +15% | 0 | 0 | — |
case-22 | fail→pass | 20,174 | 16,401 | -19% | 1 | 1 | 0% | 3,058 | 3,542 | +16% | 0 | 0 | — |
case-23 | fail→pass | 34,461 | 14,807 | -57% | 1 | 1 | 0% | 5,316 | 3,578 | -33% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +52 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.