Simulate — Digital twins · Synthetic data · What-if
Simulate is your safe-to-fail laboratory. Run a year of pricing, policy or staffing experiments overnight — without touching a real customer or breaching a single privacy budget.
Scenario throughput
1,200×
Privacy budget (ε)
<1.0
Decision lift
+11–24%
Deliverables
Everything that ships
- Customer / journey twinAgent-based simulation calibrated on 24 months of behaviour.
- Synthetic data generatorDifferentially-private synthesis for dev, test and partners.
- What-if engineScenario sweeps over price, policy, capacity, staffing.
- Counterfactual ledgerEvery decision stored with its alternative — for learning.
- Pre-production sandboxProduction-grade twin for agent, pricing and ops experiments.
Pod composition
- Simulation Scientist
- Data Scientist
- Domain SME
Example output · Scenario · price_test_q3json
{
"twin": "uk_retail_v4",
"horizon_days": 90,
"levers": { "price_delta": "+3%", "promo_intensity": 0.7 },
"outcomes": { "revenue": "+£2.1M", "churn": "+0.4pp", "margin": "+1.8pp" },
"confidence": 0.86
}Timeline
Weeks 6–12 · first twin runs by day 70
- 1Weeks 6–8Twin calibration
Fit twin on 24 months of behaviour; validate on hold-out.
- 2Weeks 8–10Synthetic + DP
Differentially-private synthesis for dev/test/partner shares.
- 3Weeks 10–12What-if engine
Sweep levers, store counterfactuals, surface decision lift.
FAQs
Things prospects ask
Is the synthetic data GDPR-safe to share?
Yes — differential-privacy ε < 1.0 by default with k-anonymity backstops; signed off by your DPO.
How accurate are the twins?
We report calibration error per cohort; typical lift attribution sits within ±3pp of A/B truth.
Commission · S5 Simulate
Stand up Simulate in Weeks 6–12.
We'll respond within one business day with a scoping note, a fixed-price outcome contract, and a named principal. Your details sync straight into our concierge queue.
- • Outcome-priced — no T&M.
- • Sovereign by default — your data, your region, your keys.
- • Wired into the Fuel Pressure gauge from day one.