VDC Variable Derivation Platform

Expert feature design that turns detailed records into interpretable variables

VDC turns our variable-design expertise into a product. Sitting between data and models, it continuously feeds high-quality features to SAM modeling and DH S1 real-time decisions. The five-dimension method is industry-neutral — banking, FMCG, pharmaceuticals or any business with time-stamped records — and 20,000+ variables come preloaded for finance.

An industry-neutral five-dimension method20,000+ variables preloaded for finance30+ business scenariosOn-premises
A smartphone lying on a blue background, with data charts and message icons floating above the screen
20,000+preloaded derived variables (finance)Interpretable, with proven KS
10thematic dimensions for individualsProfiled across 4+ dimensions
15thematic dimensions for enterprises
30+business scenarios
Derivation method

Time + product + status + value + statistic

Every derived variable combines five dimensions: over which time window, for which product or category, for records in which status, taking which value, with which statistic. The same method generates interpretable variables in bulk from detailed records in any industry.

Time+Product+Status+Value+Statistic
Every industry

One method, variables for three industries

Each example below takes one variable to show how the five dimensions apply to different industries’ data.

Banking · application scorecardSee the solution
TimeLast 6 months
ProductCredit cards
StatusRepaid on time
ValueRepayment amount
StatisticMonthly average
FMCG & marketing · member churnSee the solution
TimeLast 90 days
ProductBeverages
StatusCompleted orders
ValueOrder amount
StatisticPeriod-on-period change
Pharmaceuticals · distributor creditSee the solution
TimeLast 180 days
ProductPrescription SKUs
StatusReceivables due
ValueDays overdue
StatisticMaximum
Common data in other industriesMembership & pointsOrders & transactionsStore POSInventoryExpenses & documentsSales flowPurchasing & paymentsEquipment & logs
Preloaded finance sourcesPersonal credit reportsTaxInvoicesBank statementsScenario dataThird-party financial data
Preloaded finance library

In finance, fifteen dimensions for a fine-grained view of every customer

The finance library applies multi-dimensional design to personal credit reports, tax, invoices, bank statements, scenario and third-party financial data, yielding 20,000+ interpretable variables with proven KS. They quantify ability and willingness to repay, and trace changes in behavior before, during and after the loan, and in marketing response.

  1. 01Borrowing self-discipline
  2. 02Repayment capacity
  3. 03Leverage
  4. 04Debt growth and appetite for new credit
  5. 05Revolving-credit expansion
  6. 06Historical delinquency
  7. 07Recent signs of easing delinquency
  8. 08Unusual credit behavior
  9. 09Ratings by other institutions
  10. 10Resilience to unusual events (job loss, illness, accidents, divorce, business failure)
  11. 11Business continuity
  12. 12Status of the controlling owner
  13. 13Collection cycles and volatility
  14. 14Profit and volatility
  15. 15Customer-base migration
Where VDC fits

Between data and models

VDC cleans and derives variables from your sample, product, transaction and operational data, feeding both SAM model building and DH S1 real-time decisions.

Enterprise dataSamples · products · transactions · operations
VDCCleansing and derivation
SAM / DH S1Model building · real-time decisions
Try online

Every SAM feature, online, for up to 10,000 samples

Upload up to 10,000 records and run the whole fourteen-step pipeline — data contract to packaged scorecard, human gates included — to see for yourself whether SAM fits your business.

Up to 10,000 samplesEvery feature enabledThe full fourteen-step pipeline
A smiling professional woman seated at a window-side desk with an open laptop
Get started

Start with one business question. Let SAM deliver a model you can sign off.

Whether you are in banking, FMCG and marketing, pharmaceuticals or another industry, try SAM online first — or talk to our modeling experts about your use case.