Insurance data in Ghana is scarce. We forge it instead.
Insurers don't open their books to students and independent researchers — not for motor, not for life, not for health, not for Home & Property insurance. Every line has the same gap: no public dataset to test a model, a regression, or a thesis chapter against. Insurance is Dataset Foundry's line family built to close that gap, sub-line by sub-line — motor, life, health, and Home & Property insurance are all live today. Ghana is today's default country, not a restriction — every generator takes a country parameter, chosen per dataset rather than fixed by the line.
seed 42 — this exact seal, every time
One platform, one line at a time
Motor, life, health, and Home & Property are all live — each with the clearest, best-documented risk factors in its category to encode. Structurally realistic data, deterministic and seeded, mimicking the shape of real scarce data rather than generating noise.
Motor insurance
Available nowGhanaian regions, vehicle makes, NIC-style policy classes, GHS premiums — deterministic, seeded, ready to forge.
Generate a datasetLife insurance
Available nowMortality-linked policy and claims data, structured the same way — configurable, seeded, documented.
Generate a datasetHealth insurance
Available nowClaims frequency and severity shaped by the risk factors researchers actually need to test.
Generate a datasetHome & Property insurance
Available nowFire & Burglary risk factors — building age, construction type, security features, prior claims — with the same seed-and-cite reproducibility as the rest of the platform.
Generate a datasetRelationships tested and confirmed in every batch
These aren't hidden in the noise — they're the reason each module exists. Every dataset forges the same directional relationships between risk factors and outcome, so your regression, GLM, or classroom exercise has something real to recover.
Motor insurance — claim severity
Driver age
Smaller claims
Older drivers tend to drive more cautiously — claim size falls as age rises.
Gender
Larger claims (male)
Male drivers are linked in the literature to higher-speed, higher-impact driving.
Driving experience
Smaller claims
More experienced drivers are better at avoiding serious accidents.
Past claims count
Larger claims
Drivers with more claims history run larger claims too, not just more frequent ones.
Life insurance — claim frequency
Applicant age
Higher frequency
Older applicants sit in higher age bands on the mortality table — the anchor itself, not a separate coefficient.
Smoker status
Higher frequency
Smokers carry an elevated baseline mortality loading — the standard actuarial rule of thumb.
BMI band
Higher frequency
Overweight and Obese bands carry an elevated mortality loading relative to Normal/Underweight.
Occupation class
Higher frequency
Hazardous occupations (mining, construction, haulage) carry a higher mortality loading than office/professional roles.
Health insurance — claim frequency
Applicant age
Higher frequency
Older applicants sit in higher age bands on the claim-incidence table — the anchor itself, not a separate coefficient.
Chronic condition
Higher frequency
A diagnosed chronic condition carries an elevated utilization loading, reflecting routine monitoring and complication-related care.
Smoker status
Higher frequency
Smokers carry a modest utilization loading for care outside what's already captured by chronic-condition status.
BMI band
Higher frequency
Overweight and Obese bands carry an elevated utilization loading relative to Normal/Underweight.
Home & Property insurance — frequency & severity
Building age Frequency
Higher frequency
Older buildings claim more often — electrical and plumbing systems degrade with age.
Security features present Frequency
Lower frequency
Alarm/CCTV/guard presence deters burglary claims specifically, which make up most of this line's claim volume.
Construction type Severity
Larger claims (combustible)
Wood/Thatch and Mixed construction burns and floods worse than Brick/Concrete/Sandcrete, running larger claims when one occurs.
Prior claims count Severity
Larger claims
A property with a worse claims history tends to run larger subsequent claims too, not just more frequent ones.
Configurable length, configurable variables
Pick a record count and a seed, and a deterministic model — frequency of claims, then severity of the ones that happen — runs entirely client-side and hands you a CSV. No accounts, no uploads, no server round-trip.
Plain-language column customization, on a paid tier
The roadmap adds more lines of insurance, a backend for larger batches and saved presets, and column customization you can describe in plain language instead of editing coefficients by hand — still aimed at mimicking real, scarce insurance data, not producing random data for its own sake. Full detail is in the roadmap.
Other countries' insurance markets are on the roadmap too, opt-in per generation — Ghana stays the default, not a ceiling.