Practice

Banking

Credit decisioning, pricing, treasury and finance integration rebuilt as governed, instrumented data products. Run by people who have priced and validated mortgage and unsecured books at Tier-1 South African banks. (Risk-management work has its own practice — see Risk Management.)

Banking Bank-specific decisioning and finance. Top: a credit-decision pipeline — application to priced offer to booked decision — with a single accent step on the priced offer. Bottom: three KPI tiles — net interest margin, loan book and finance close — reconciling across the bank. application scored priced decision moment booked CREDIT DECISIONING · ONE LINEAGE NIM 3.62% LOAN BOOK R 218bn FINANCE CLOSE D + 4 Banking Bank-specific decisioning and finance. Top: a credit-decision pipeline — application to priced offer to booked decision — with a single accent step on the priced offer. Bottom: three KPI tiles — net interest margin, loan book and finance close — reconciling across the bank. application scored priced decision moment booked CREDIT DECISIONING · ONE LINEAGE NIM 3.62% LOAN BOOK R 218bn FINANCE CLOSE D + 4

Practice signals 1 / 3

Decisioning needs a governed route

Credit, customer and risk signals must move through a route that can be explained.

Who this is for

Heads of retail and corporate credit, treasury, finance, MI and product at banks and bank-adjacent businesses.

Where we help

Most of a bank's day-to-day decision systems still run on a patchwork of operational extracts and finance spreadsheets. Pricing engines and credit policy live in one set of tools; finance and MI live in another; treasury sits between the two. Reconciling them takes longer than the decisions they support, and most of that effort produces no defensible audit trail.

What we do

  • Re-engineer credit decisioning and pricing pipelines as governed, versioned data and modelling products — separated from the back-office reporting that consumes them.
  • Productionise treasury and ALM operational data flows — funding, cash forecasting, intercompany and behavioural inputs.
  • Build the integration layer between core banking, finance and risk that most banks still cobble together each cycle.
  • Modernise bank-specific MI: branch, product, segment and channel performance, traceable to source.
  • Document data and model lineage to a standard that holds up under PRA / Prudential Authority review.
IFRS 9 ECL portfolio dashboard mockup An IFRS 9 ECL dashboard summary. Three KPI tiles across the top — total ECL, coverage ratio, stage 3 share — followed by a stage-share bar and a 12-month ECL trend line. One stage-3 portfolio tile is flagged in accent red as the explainable outlier. IFRS 9 ECL — RETAIL CREDIT PORTFOLIO MOCKUP A governed BI data product that carries decisions TOTAL ECL R 4.2bn ↓ R 180m vs Q4 2025 COVERAGE RATIO 3.4% unchanged · within RAS STAGE 3 SHARE 7.8% ↑ 80bps · explainer ready RECONCILED TO Finance close · Risk MI Regulatory FINREP all three agree EXPOSURE BY IFRS 9 STAGE — R bn Stage 1 · 75.0% Stage 2 · 17.2% S3 · 7.8% ECL TREND · 12 MONTHS Apr 25 Jul 25 Oct 25 Apr 26 EXPLAINER FOR S3 ↑ 80BPS Driver — Motor portfolio · 14 obligors moved S2 → S3 Macro overlay refreshed (Mar 2026 PD curves) Traced back to source PD model run · auditable IFRS 9 ECL portfolio dashboard mockup An IFRS 9 ECL dashboard summary. Three KPI tiles across the top — total ECL, coverage ratio, stage 3 share — followed by a stage-share bar and a 12-month ECL trend line. One stage-3 portfolio tile is flagged in accent red as the explainable outlier. IFRS 9 ECL — RETAIL CREDIT PORTFOLIO MOCKUP A governed BI data product that carries decisions TOTAL ECL R 4.2bn ↓ R 180m vs Q4 2025 COVERAGE RATIO 3.4% unchanged · within RAS STAGE 3 SHARE 7.8% ↑ 80bps · explainer ready RECONCILED TO Finance close · Risk MI Regulatory FINREP all three agree EXPOSURE BY IFRS 9 STAGE — R bn Stage 1 · 75.0% Stage 2 · 17.2% S3 · 7.8% ECL TREND · 12 MONTHS Apr 25 Jul 25 Oct 25 Apr 26 EXPLAINER FOR S3 ↑ 80BPS Driver — Motor portfolio · 14 obligors moved S2 → S3 Macro overlay refreshed (Mar 2026 PD curves) Traced back to source PD model run · auditable

Diagram signals 1 / 3

Decisioning needs a governed route

Credit, customer and risk signals must move through a route that can be explained. The body diagram should make this route explicit enough to discuss in a working session.

Credit decisioning pipeline from application to booked, with one lineage.

Outcomes

  • Pricing and credit decisions that reconcile to finance and MI without manual stitching.
  • A treasury and ALM data layer that the front office and finance both trust.
  • Faster month- and quarter-end finance close on the bank side.
  • Clear lineage from source transaction to disclosed number.

Engagement model

We typically start with a focused assessment of one decision pipeline — pricing, credit policy, treasury cash forecasting or finance close — and expand outward. Risk-management work (capital, IFRS 9, market and liquidity risk, MRM) is delivered through our Risk Management practice.