What makes the work necessary
IFRS 9 joins accounting policy, credit risk methodology, portfolio data and repeatable calculation. Weakness in any one layer can make Expected Credit Loss results difficult to explain or reproduce.
Signals that the current approach needs attention
Definition of default and staging rules are inconsistently applied
PD term structures and forward-looking adjustments are difficult to reproduce
ECL calculation depends on manual spreadsheet steps
Monitoring and governance evidence is fragmented
Capabilities the engagement is designed to create
- Traceable calculation from source data to ECL output
- Consistent staging, parameter and scenario logic
- Controlled, repeatable calculation pipeline
- Clear methodology, controls and monitoring requirements
Tangible outputs for analysis and handover
Structured from framing to transfer
Each stage leaves an output that can be reviewed before the next stage begins.
Methods selected around the analytical purpose
The tool follows the control environment
Evidence needed to support the scope
- Exposure, contractual cash flow and risk parameter data
- Default, recovery and behavioural history where modelling is in scope
- Macroeconomic series and approved scenario assumptions
Controls that preserve accountability
- Policy choices and expert judgement are explicitly documented
- Reconciliation and control totals are retained with each run
- Monitoring thresholds and escalation responsibilities are defined
Source-verified implementation examples
Before scoping the work
Can you modernise an existing spreadsheet ECL process?
Yes. The calculation logic, inputs, controls and outputs can be mapped before moving repeatable steps into a controlled Python, R or VBA implementation.
Does the service include model monitoring?
Monitoring can cover data drift, staging movement, parameter performance, scenario effects, reconciliation and documented review thresholds.