Case studies

What changed after the analysis?

The full stories: awkward starting points, methods that survived contact with reality, and the numbers that moved afterwards.

Illustrative machine-learning lead scoring tier output
TelecommunicationsThailand

Transforming Customer Targeting, 3x Conversion Rate Improvement

How True Corporation moved from gut-feel segmentation to ML-driven targeting — and tripled conversion rates in weeks.

conversion rate improvement
Forecast risk profile across a 320-store retail estate
Specialty retailWestern Europe

320 Stores. One Number Each. Finally One They Could Trust.

How a European Specialty Retailer Replaced Estate-Wide Spreadsheet Guesswork with Store-Level Revenue Forecasts, Confidence Bands, and a Planning Cadence That Actually Held.

9.8%chain MAPE, down from 24.3%
Manufacturing scrap rate by tool age from causal analysis
ManufacturingEurope

4% Scrap. The Dashboards Said Everything. They Explained Nothing.

How a European Manufacturer Stopped Guessing at Root Causes and Used Graphite Note Desicion Lever framework to Cut Scrap by 1.9 Percentage Points in Six Months.

4.1% → 2.2%scrap rate in six months
Forecast accuracy drift from untracked planner overrides over time
Automotive distributionCentral America

From 39% Forecast Error to Decision-Ready Demand Intelligence

How a mid-market automotive parts distributor replaced ERP-default forecasting with a 50,000-SKU decision intelligence system

39% → 8%forecast error (MAPE)
Guest behavioural segment and property preference audience heatmap
Hospitality & tourismEurope

Guest Intelligence at Scale: Four-Dimension Audience Segmentation for a Major Resort Group

How Decision Intelligence gave a multi-property resort operator its first complete view of 3.8 million guests — without touching the CRM query interface.

3.8M+guest accounts segmented
Customer probability-alive decay curves by churn-risk segment
Omnichannel retailUnited States

From Gut Feel to CLV Score: How One Retailer Stopped Guessing Which Customers Were Worth Keeping

How a US Omnichannel Home Goods Retailer Used CLV Modelling to Identify At-Risk Customers Before They Churned, Score Every Buyer by Future Value, and Shift $4.1M in Marketing Spend to Where It Would Actually Return.

$4.1Mmarketing budget reallocated

What decision is holding your team back?

Bring us the business question and the data you already have.