How Scores Work
Scoring turns a country's raw facts into a single comparable number for a specific question. The pipeline is deliberately simple so it can be audited: normalize each input, weight it, sum the weighted parts. Nothing else enters the calculation.
Last updated:
Quick answer
Each input is normalized to a 0–100 sub-score, multiplied by its published weight, and summed — the total is the country's score for that ranking, fully reproducible from the country data.
Step 1 — Normalize each input
Each factor is converted to a 0–100 sub-score where 100 is most favourable for the question being asked. Rates where lower is better are inverted; difficulty scales are converted to ease; availability booleans map to endpoints. This removes unit mismatches so factors are comparable before weighting.
Step 2 — Apply published weights
Each normalized sub-score is multiplied by the factor's weight from the ranking's published table. Weights sum to one, so the weighted parts sum to a 0–100 total. A higher weight means that factor moves the result more — which is exactly what the weight is meant to express.
Step 3 — Sum and order
The weighted sub-scores are summed into the country's score, and countries are ordered by that score. Because every step is arithmetic over published inputs, two readers with the same data will always reproduce the same order.
| Factor | Sub-score (0–100) | Weight | Contribution |
|---|---|---|---|
| First-party Stripe | 100 | 0.40 | 40.0 |
| EU/EEA market access | 100 | 0.25 | 25.0 |
| Formation ease | 75 | 0.20 | 15.0 |
| Compliance simplicity | 60 | 0.15 | 9.0 |
| Total | — | 1.00 | 89.0 |
Worked example — a payments-weighted score (illustrative weights)
How GeoBusinessIQ calculates a ranking
Read typed country data
Start from the same validated country facts shown on every profile.Normalize each factor
Convert rates, difficulty scales, and availability to a common 0–100 sub-score.Apply published weights
Multiply each sub-score by the factor's weight from the ranking's published table (weights sum to one).Sum and order
Add the weighted sub-scores into the country's score and order jurisdictions by it — fully reproducible.
Methodology notes
- The worked example uses illustrative weights to show the arithmetic; each ranking's real weights are published on its page and in the rankings methodology.
- Scores are relative within the covered country set, not absolute grades — a score of 89 means 'strong on these weighted factors among covered jurisdictions', not '89% ideal'.
Data limitations
- Scores compare only the factors in the scorer; they cannot represent factors outside the dataset.
- Sub-score normalization is relative to the covered range, so adding countries can shift sub-scores.
FAQ
- Can I reproduce a country's score by hand?
- Yes. Take the country's factor values, normalize each as described, multiply by the published weights, and sum. The total is the score used to order the ranking.
- Is a higher score always 'better'?
- Better for the specific question the ranking asks, among the covered countries. It is not a universal quality grade and does not account for your business-specific factors.
Related
Sources
- OECD — OECD — economic and tax statistics (accessed ; reviewed )Covers: Comparable corporate tax, statutory rate, and economic indicators across member and partner economies.Does not cover: Effective tax rates, deductions and incentives, local surtaxes, and personal residency rules.Why it matters: Used as a cross-country baseline to sanity-check rates against primary tax-authority figures.Review cadence: Annual, plus on major statutory changes.
- European Commission — European Commission — policy and country information (accessed ; reviewed )Covers: EU policy framework including the VAT One-Stop-Shop and single-market rules.Does not cover: Member-state-specific reduced rates, national thresholds, or non-EU jurisdictions.Why it matters: Used for EU/EEA market-access and VAT-OSS framing referenced across rankings and guides.Review cadence: On policy change; re-checked each data review.
Last updated: