Strategic Workforce Planning: Re-shaping, Not Hiring
- Maturity
- L3
- Domain
- Plan & Cost
- Analytics
- strategic
- Workforce cost-income contribution
- ≈34pp
- Sponsor
- Group Chief Operating Officer
- Confidence
- Moderate
The situation
What must Aurora's workforce become — by segment, role and skill — as AI re-prices tasks and the cost-income target tightens, reconciled into a redeployment and role-redesign plan rather than a hiring or redundancy plan?
The recommendation on the table
Adopt BNK-01 as the single reconciled re-shaping baseline
Gives every redeployment, control-coverage and cost-income decision one trusted view of roles, task content and AI exposure across five segments.
Trade-offRequires reconciling HR position data with finance, a role/task taxonomy and a control-obligation register — a one-off integration across system owners.
The evidence
Aurora's board has committed to a cost-income ratio below 55% and a bank-wide AI programme — both, in substance, workforce commitments, since roughly half the controllable cost base is people. Yet Aurora could not produce one reconciled view of what its 96,500-person workforce actually does, how exposed each role is to AI, or whether its control functions are staffed for an AI-augmented bank. BNK-01 built that baseline across five segments and ~45 countries and reframed planning from a hiring question into a redeployment-and-redesign question: not how many to hire, but how many to move, reskill and redesign before their work is automated.
Aurora — Strategic Workforce Planning
The reconciled re-shaping baseline: where AI exposure lands, where control coverage is thin, and how agile the workforce really is.
| Low | Medium | High | |
|---|---|---|---|
| Operations | 1 | 2 | 4 |
| Analytical | 2 | 4 | 2 |
| Technology | 2 | 3 | 2 |
| Advisory | 4 | 2 | 1 |
| Control | 3 | 3 | 1 |
Key takeawayExposure concentrates in operations — the largest population, the clearest redesign target.
Key takeawayOperations drives the largest people-cost share of the ratio while carrying the highest exposure.
Key findings
Operations & processing — about 21,200 people, a fifth of the bank — contributes disproportionately to the cost-income ratio and sits at 55–65% AI task-exposure. It is simultaneously the clearest redesign target and the largest redeployment challenge, and today there is no role-redesign plan in place for it.
What we can’t claim
Aurora has committed publicly to a skills-based, agile workforce, but only ~12% of the workforce is in skills-based deployment and the mobility-velocity data to run redeployment barely exists. The bank is planning to move people out of automated roles at a pace its own data says it cannot yet achieve.
Recommendations
Adopt BNK-01 as the single reconciled re-shaping baseline
high priorityGives every redeployment, control-coverage and cost-income decision one trusted view of roles, task content and AI exposure across five segments.
Trade-off
Requires reconciling HR position data with finance, a role/task taxonomy and a control-obligation register — a one-off integration across system owners.
Sequence redeployment and reskilling ahead of automation for high-exposure operations
high priorityCaptures AI value without the conduct and morale cost of cuts, and protects control coverage as roles change.
Trade-off
Front-loads reskilling investment before automation savings land; demands disciplined planning.
Analytical framework
How we reached this
Strategic diagnostic — reconcile the workforce and its AI task-exposure to drive a redeployment, control-coverage and cost-income plan.
ConfidenceMedium-High
Analytical framework
How we reached this
Strategic diagnostic — reconcile the workforce and its AI task-exposure to drive a redeployment, control-coverage and cost-income plan.
Methods applied
Statistical techniques
Algorithms
Data sources
Outputs generated
Why this confidence
Permanent data is reconciled and solid; AI task-exposure is a structured, rubric-based estimate with a stated band, which caps confidence below High.
The reasoning
Business context
The foundational Banking project, sponsored by the Group COO rather than HR — a deliberate signal that workforce planning here is an operating-model decision. It reconciles HR-owned position and role data with finance cost/income, a role/task taxonomy, contingent reliance and a control-obligation register, the joins a bank rarely makes.
Expected value
A reconciled workforce-and-exposure baseline is the prerequisite for everything downstream: productivity (BNK-02), AI transformation (BNK-03) and the digital twin (BNK-05) all consume it. Concretely, BNK-01 sizes redeployment-versus-redundancy savings, structural contingent, and the control-coverage gap that automation would otherwise widen.
Workforce landscape
Operations & processing (≈21,200 staff) carries both the heaviest cost-income contribution and the highest AI exposure. AI-model-governance roles are nascent (<50% staffed). Group agility (~12%) is far below the level the stated operating model assumes — the mobility to redeploy people out of automated roles is not yet evidenced in the data.
The analytics journey
Level 3, strategic. Descriptive and diagnostic by design: it reconciles the workforce, maps AI task-exposure as a structured assessment, and frames the redeployment choice — without predictive modelling. Honest about its soft inputs: AI task-exposure is a structured estimate with a stated band, and agility data is incomplete.
Under the hood
A reconciliation model nets roles, task content and AI exposure against demand and the cost-income path; a tenure-based rule separates structural from variable contingent; control coverage counts qualified, in-role staff against the obligation register. The AI-exposure layer is rubric-based and transparent (defensible to regulators), not a black-box score.
Confidence & evidence
Why you can rely on this
The inconvenient truth
Aurora has committed publicly to a skills-based, agile workforce, but only ~12% of the workforce is in skills-based deployment and the mobility-velocity data to run redeployment barely exists. The bank is planning to move people out of automated roles at a pace its own data says it cannot yet achieve.
Method
Confidence is a deterministic read of KPI strength, target and benchmark coverage across this project — shown on an illustrative reference dataset, computed the same way it would be on live data.
Take this further