People Centricity case study

Choose a Pool or a Plan Before You Design Succession

280real positions used to test the criticality grid
77%of positions classified the same with fewer than half the criteria
42activities, each with one owner, in the succession process
27indicators, each with an owner, from mapping to retention

The situation

The question behind the question.

Succession was one of the initiatives of this institution’s six-year people strategy, and time was short. A quarter of its senior managers were due to retire within 5 years, and in some departments every head of service would change within 3. The labour market offered little cover: the country expects to be short of tens of thousands of nurses and doctors by 2040. Its talent team had already built a criticality grid and run nursing pilots. It lacked a model that would hold across a matrix of directorates and departments, and proof its grid measured the right things.

As it was asked

“Review our approach and propose a governance model for succession.”

As we reframed it

Test the criticality grid before redesigning anything, learn from peers and from our own pilots, then give succession one process, 2 routes and one owner for every step in the matrix.

How we worked

How the work unfolded.

01

Test the grid

We started with the evidence the team already had. Its grid scored 9 criteria; we tested 4 on 280 real positions from 14 medical and nursing units: the impact of the post and of its vacancy, the rarity of its skills, and the time a newcomer needs to work alone. Each was scored 1 to 3 against anchors in months: a post refilled within 3 months scores 1, one that takes a year or more scores 3. With fewer than half the criteria, 77% of positions landed in the same class, a correlation of 0.89.

02

Look outside and back

Then 2 sources of lessons. Some 45 organisations were approached and the closest peers interviewed in depth: a university hospital group, a technical university and 2 national employers, all steering from the centre while managers own the outcome. The institution’s own nursing pilots were then scored step by step with their owners: mapping positions and supporting development were weakest, and the tools would not scale.

03

Design the governance

Then the process, worked through with the HR business partners and the talent team: 5 steps from mapping to recruiting, and one choice at the start, a talent pool where roles share skills, a career plan where a role is unique. A nurse manager answers to a department and to the nursing directorate; without a rule, succession belongs to both, so to neither. A 42-activity matrix gave every step one owner across that double line. The top band was raised and crossed with how soon a post falls vacant, and each successor got a readiness level and a development plan.

04

Plan the scale-up

Then the scale. Succession covered about 7% of positions, and only on request. 3 scenarios to 2030 raised coverage to a quarter, a half or every critical post, each weighed against the central team’s workload and the training budget. 27 indicators, each with an owner and a 6-month rhythm, ran from positions mapped to critical posts with no successor. The spreadsheet had reached its limits, so the system requirements were set out too.

The instrument

Pool or plan

A pool. Where several roles share the same skills, develop a group of candidates, any of whom can step in.

A plan. Where a role is unique, name its successors, give each a readiness level and a development plan.

The choice. Made at mapping, by the department with HR, before anyone is named.

The pool develops people for families of roles; the plan protects the few roles no one else can do. Choose first, and the process follows.

What moved

What changed, and what didn't.

4criteria now rate every key position, down from 9
Evidence Decision taken Adopted in practice Measured impact
What moved

The institution now rates its key positions on 4 criteria instead of 9, prioritised by how soon each will fall vacant. The governance runs as designed: the talent service guards the method, HR business partners lead, departments shortlist through their career committees, and both routes, talent pools and career plans, are in use.

What didn't move

When we handed over, no scale-up scenario had been chosen and most of the 27 indicators were not yet produced. The pilots filled every succession internally, leaving the balance with external hiring open, and the system that would replace the spreadsheet was still to come. Future importance was dropped as a criterion, so the grid reads today’s roles better than tomorrow’s.

What we'd say to a peer

Before you redesign succession, test the grid you already have: if 4 criteria give the same answer as 9, keep 4 and spend the effort on what follows. Then build pools where roles share skills, career plans where a role is unique, and give every step one owner across the matrix.

Let's talk Discover People Centricity Read more cases
Method note:

Own approach in 4 steps; 9- versus 4-criterion test on 280 positions; external benchmark; pilot lessons learned; 5-step process with 2 routes; 42-activity responsibility matrix; 27 indicators; deployment scenarios.

WeAreHuman@Work

One piece of evidence on sustainable work, every Sunday. Free, and one click to leave.

Get the weekly edition