Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply the Born Global framework to analyze firms that internationalize rapidly from inception under resource constraints. Use this skill when the user needs to evaluate whether a startup or SME can pursue early internationalization, identify the capabilities enabling born globals, or design a resource-constrained international market entry strategy.
.claude/skills/asgard-ai-platform-grad-born-global/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 33% | 0% |
Born globals are firms that internationalize at or near founding, achieving significant foreign sales (typically 25%+ of revenue) within 3 years of inception. They challenge the Uppsala model's assumption of gradual, stage-based internationalization. Born globals succeed not through scale or accumulated experience, but through knowledge intensity, network relationships, and focus on global niche markets. They are most common in knowledge-intensive, technology-driven industries.
Trigger conditions:
When NOT to use:
IRON LAW: Born Globals Succeed Through Knowledge Intensity
and Network Relationships, NOT Scale
A born global cannot out-resource an MNC. Its advantages are:
1. Unique knowledge or technology (hard to imitate)
2. Network relationships that provide market access and legitimacy
3. Niche focus that large firms find unattractive
If the firm lacks ALL THREE, early internationalization is premature
and will likely fail. Scale-dependent businesses cannot be born global.Evaluate whether the firm meets born global preconditions:
| Factor | Required for Born Global | Assessment | |--------|-------------------------|------------| | Knowledge intensity | High — proprietary technology, IP, or expertise | Yes / No | | Founder international experience | Prior work/study abroad, multilingual, global network | Yes / No | | Niche market orientation | Product serves a global niche too small for MNCs | Yes / No | | Digital/scalable delivery | Product can be delivered internationally at low marginal cost | Yes / No | | Home market limitation | Domestic market too small to sustain the business | Yes / No |
If fewer than 3 factors are present, the gradual Uppsala path may be more appropriate.
Born globals rely on networks to overcome resource constraints: personal (founder contacts), industry (trade associations), institutional (government programs, accelerators), and customer (referral chains). Map which relationships provide market access, credibility, or knowledge.
Born globals cannot enter all markets. Prioritize based on:
Born globals typically use low-commitment, high-reach modes:
markdown# Born Global Assessment: {Firm} ## Eligibility Check | Factor | Present? | Evidence | |--------|----------|----------| | Knowledge intensity / Founder intl exp / Niche / Digital delivery / Home market limit | Yes/No | {details} | ## Verdict: Born Global Viable / Not Viable ## Network Map - {Network type}: {key relationships and what they provide} ## Target Markets (prioritized) 1. {Market}: {rationale} | Entry mode: {mode} | Timeline: {milestones}
references/born-global-criteria.mdreferences/inv-framework.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 55,567 | 22,611 | -59% | 1 | 1 | 0% | 4,456 | 4,159 | -7% | 0 | 0 | — |
case-02 | fail→pass | 33,403 | 16,383 | -51% | 1 | 1 | 0% | 5,022 | 3,870 | -23% | 0 | 0 | — |
case-03 | fail→fail | 47,673 | 32,006 | -33% | 1 | 1 | 0% | 6,341 | 5,284 | -17% | 0 | 0 | — |
case-04 | pass→pass | 25,463 | 21,184 | -17% | 1 | 1 | 0% | 3,201 | 4,249 | +33% | 0 | 0 | — |
case-05 | pass→pass | 28,713 | 27,534 | -4% | 1 | 1 | 0% | 4,265 | 5,502 | +29% | 0 | 0 | — |
case-06 | pass→pass | 23,226 | 29,512 | +27% | 1 | 1 | 0% | 3,223 | 4,735 | +47% | 0 | 0 | — |
case-07 | pass→pass | 18,262 | 12,384 | -32% | 1 | 1 | 0% | 2,736 | 3,017 | +10% | 0 | 0 | — |
case-08 | pass→pass | 30,086 | 18,049 | -40% | 1 | 1 | 0% | 2,998 | 3,692 | +23% | 0 | 0 | — |
case-09 | pass→pass | 20,598 | 19,427 | -6% | 1 | 1 | 0% | 3,093 | 4,059 | +31% | 0 | 0 | — |
case-10 | pass→pass | 19,182 | 17,621 | -8% | 1 | 1 | 0% | 2,587 | 3,300 | +28% | 0 | 0 | — |
case-11 | pass→pass | 20,898 | 22,431 | +7% | 1 | 1 | 0% | 2,437 | 3,206 | +32% | 0 | 0 | — |
case-12 | pass→pass | 17,455 | 33,928 | +94% | 1 | 1 | 0% | 2,640 | 3,711 | +41% | 0 | 0 | — |
case-13 | pass→pass | 19,112 | 20,524 | +7% | 1 | 1 | 0% | 2,322 | 3,672 | +58% | 0 | 0 | — |
case-14 | pass→pass | 16,193 | 18,605 | +15% | 1 | 1 | 0% | 2,522 | 3,917 | +55% | 0 | 0 | — |
case-15 | fail→fail | 15,413 | 10,747 | -30% | 1 | 1 | 0% | 2,657 | 2,763 | +4% | 0 | 0 | — |
case-16 | pass→pass | 15,154 | 14,085 | -7% | 1 | 1 | 0% | 2,263 | 2,709 | +20% | 0 | 0 | — |
case-17 | pass→pass | 16,417 | 21,908 | +33% | 1 | 1 | 0% | 2,380 | 4,444 | +87% | 0 | 0 | — |
case-18 | fail→fail | 25,400 | 16,561 | -35% | 1 | 1 | 0% | 3,190 | 3,831 | +20% | 0 | 0 | — |
case-19 | fail→pass | 42,626 | 18,650 | -56% | 1 | 1 | 0% | 5,580 | 3,415 | -39% | 0 | 0 | — |
case-20 | fail→pass | 18,809 | 15,017 | -20% | 1 | 1 | 0% | 2,412 | 3,005 | +25% | 0 | 0 | — |
case-21 | pass→pass | 18,921 | 17,666 | -7% | 1 | 1 | 0% | 2,407 | 3,719 | +55% | 0 | 0 | — |
case-22 | pass→pass | 15,326 | 14,473 | -6% | 1 | 1 | 0% | 2,308 | 3,343 | +45% | 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. 22 cases were attempted. The headline lift of +18 percentage points is the difference between those two pass rates over the 22 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.