How to use a skills matrix in a restructure

Content aligned to the Capability Guide PDF for this topic. Q2 2026 refresh.

How do you restructure on capability instead of the org chart?

Restructures decided on boxes and reporting lines answer the wrong question. The chart shows who reports to whom; it does not show who can do the work the new design demands. CIPD Labour Market Outlook shows many UK employers still report hard-to-fill vacancies linked to capability — shedding people you will re-hire is expensive when skills data could redeploy them instead (Chartered Institute of Personnel and Development, 2024).

A skills matrix supplies the missing fact base: real capability per person, so you place people into new roles on evidence, retain transferable talent, select fairly where headcount must fall, and see gaps before day one of the new structure.

What changes when reorganisation uses skills data?

Reorganise on capability, not titles. Reporting lines and tenure say little about fit in new roles.

Redeploy, do not only remove. Transferable skills show who could move into the customer hub, specialist unit, or delivery squad — retaining knowledge you would otherwise pay to rebuild.

Build fair selection pools. Where roles must reduce, group people doing similar work at similar capability with transparent criteria.

See the new structure's gaps. Mapping people to new design exposes capabilities the org will lack — hire or develop before go-live.

This guide complements Before you restructure on this site; that page frames the decision — this page covers running the matrix through the move.

Which four questions must a restructure answer?

• Who fits the new roles? Match capability to each new role's requirements — confident move, stretch with plan, or not a fit.
• Who can we redeploy? People one level short or with adjacent skills — cheaper and kinder than redundant-then-rehire.
• How do we select fairly? Pools of similar capability and documented criteria when headcount reduces.
• Where will we be short? New design minus current capability — gaps to fill before customers feel them.

What are the seven steps to run a skills-based restructure?

1. Inventory current capability. Score the workforce on one scale before rumours harden — dated evidence beats memory under stress.
2. Define new roles by skills and levels. Each new box gets required skills, not only a title.
3. Map matches and stretches. At or above required = confident; one below = redeploy with development; two below = different role or exit path.
4. Prioritise redeployment. Thick blue ribbons in a move diagram — most people map cleanly when capability leads.
5. Form selection pools where needed. Similar work, similar scores, same process — legal and cultural defensibility.
6. Plan gap closure. Specialist unit short on security? Hire or develop before launch.
7. Communicate with evidence. People deserve to know why a match fits — reduces poisoned launches.

What does a restructure move pattern look like?

Illustrative consolidation — current teams into new structure (34 people):

What the restructure lead reads. Most people map cleanly on evidence — not guesswork. Green stretch moves into the specialist unit need development plans, not rejection. Only three lack any fit — concentrate the hardest human conversations there instead of across the whole workforce. Right-hand nodes confirm each new team is staffed; remaining holes are visible for hire or upskill.

Worked placement. New role needs L3 customer handling and L3 data · Person X at 3 and 3 → confident · Person Y at 3 and 2 → stretch, develop data · Person Z at 1 and 1 → not this role — find another channel.

How do you score fit for new roles on 0–5?

Weightings (25%–100%) let you express fit as a percentage when comparing two placement options for the same role.

How do you keep selection defensible?

Where headcount must fall, document pools: role family, capability band, selection criteria applied to all. The matrix does not make the call — it grounds the call in evidence everyone can see. Pair with HR and legal on local rules; capability data supports objective treatment, not replacement for process.

Avoid using the matrix as a secret ranking employees never see. Transparency in criteria beats surprise in outcomes.

What mistakes turn restructures into talent drains?

Chart-only design. New boxes without skill requirements.

Assuming title implies fit. Support lead ≠ customer hub lead without scores.

Redundant-then-rehire. Losing capability you need six months later.

No gap plan. Launching structure short on specialists.

Stale scores. Basing moves on pre-pandemic ratings.

Mixing performance and capability. Restructure placement is about can do, not last year's appraisal narrative.

What if unions or consultation require role comparison?

Edge case: collective consultation. Export anonymised capability distributions per pool — not public individual scores — to show how groups were defined. Agree which skills are in scope for selection versus redeployment-only tracks. Keep individual records for appeals with assessor notes and dates.

When two countries merge operating models, harmonise descriptors before comparing scores — a 3 in one region must mean the same as a 3 in another.

When should scoring happen relative to consultation?

Baseline capability before design leaks — rumours distort scores if people perform anxiety instead of skill. Refresh after consultation milestones when role definitions stabilise. Do not score during active grievance on unrelated matters without HR guidance. Document assessor and date for every cell used in placement or selection.

How do you model the new structure's workload?

Capability fit is necessary, not sufficient. After mapping people, sense-check capacity: headcount per new node versus work volume. A hub staffed with confident matches still fails if spans are impossible. Pair matrix output with workload data — the matrix answers who can do the work; capacity planning answers how much.

What happens after day one?

Rescore at 90 days in new roles. Stretch placements should show movement toward required levels; gaps in the new structure should drive hiring or learning plans. Link to onboarding for people entering new roles and development plans for stretch moves.

How do you map people into the new design step by step?

Step one: freeze new role profiles with required skills and levels. Step two: refresh current scores with calibration — restructure is where generous ratings hurt most. Step three: generate a fit table — each person against each new role they could credibly take, marking confident, stretch, or non-fit. Step four: assign confident matches first, freeing stretch candidates for roles that need them. Step five: redeploy stretch with written plans — skill, target level, support, review date. Step six: address non-fits through alternate roles, pools, or exit pathways per policy. Step seven: list remaining capability holes in the new structure for hire or develop.

Gartner's finding that few organisations hold reliable skills data explains why many restructures repeat the same placement arguments — a shared matrix ends the circular debate (Gartner, 2024). World Economic Forum figures on skills change remind you to score against future role bars, not only yesterday's job descriptions (World Economic Forum, 2025).

How do you communicate capability moves with dignity?

Separate scores from personal worth. Explain that levels describe fit against future role requirements. Offer stretch moves as investment with plans; confident moves as recognition of readiness. Non-fit conversations stay factual — which requirements are unmet — without labels about the person.

Employees who see their own row — where policy allows — can self-identify redeployment options managers missed. That transparency reduces rumours that restructuring is arbitrary when the grid is shared appropriately.

Keep a decision log: role, person, match type, evidence date, development plan if stretch. Auditors and employee representatives ask for process — the matrix supplies substance behind each placement without replacing legal review.

When the new structure launches short on a capability, treat it as a programme: hire, contractor, or develop — named owner and date — rather than hoping the team absorbs the gap informally.

Scenario planning helps: run two future org designs against the same capability data — "hub model" versus "regional model" — and compare how many confident matches each creates before you announce. Restructures that look elegant on paper fail when the matrix shows only twelve confident fits for twenty new roles.

HR information systems hold titles; matrices hold capability. Integrate them by linking person IDs, but never replace scored skills with job codes alone — codes lag reality by months in fast-changing teams.

Leaders sometimes fear scores will demoralise — the alternative is opaque decisions that demoralise more. Sharing fit types (confident, stretch, exploring) before final announcements, with development offers for stretch, preserves dignity while keeping evidence central.

Post-restructure, measure success by whether required capabilities in the new design reach target levels within two quarters — not only whether the chart was published on time. The matrix becomes your early warning if the new structure is under-skilled in practice.

Archive the pre-restructure matrix version — you may need to explain placement decisions months later. Version numbers and dates on exports are part of defensible process, not bureaucracy. Treat that archive like financial records: stored, retrievable, owned, and reviewed when disputes arise. That discipline protects both the organisation and the people affected.

How do unions and works councils use capability data?

Provide criteria, not raw rankings, in consultation. Works councils often welcome transparent skill definitions when redundancies are unavoidable — ambiguity breeds dispute more than numbers do. which skills define pools, how levels were assessed, appeal route for factual score disputes. Capability data supports objective treatment; it does not replace dialogue. Where consultation agreements require skills assessments, the matrix becomes the shared language — reduces ad-hoc comparisons that feel personal.

What does placement scoring look like skill by skill?

Headcount moves are easy to draw on an org chart; capability fit is not. Below, one candidate for a new Customer Hub Lead role — required levels in the second column.

What the restructure lead reads. Person X is a confident match on every bar — move with minimal development plan. Person Y is a stretch on data only: place with a ninety-day plan to reach Level 3 on reporting, not as a full match on day one. Person Z fails the role profile — redeploy to a channel that needs L1–L2 customer handling, not forced into the hub lead box because of tenure.

Weighting each skill (for example customer L3 = 30%, data L3 = 25%, coaching L4 = 30%, compliance L3 = 15%) lets you express fit as a percentage when two candidates are close — useful when consultation requires transparent comparison.

How do you communicate evidence-based moves without breaking trust?

People accept hard outcomes more often when they understand the criteria. Share the role skill profile (not everyone's private scores) and explain match type: confident, stretch with plan, or alternative channel. For stretch placements, name the one or two skills and the support offered — mentor, course, supervised period — so the move reads as investment, not demotion by stealth.

Where headcount reduces, document pools: role family, capability band, selection criteria applied uniformly. The matrix supports the pool definition; it does not replace consultation timelines or legal advice. Archive assessor notes and dates for appeals — "Level 3 on customer handling per descriptor v2.1, evidence: six months unsupervised queue ownership."

What happens in the first ninety days after day one?

Rescore at day 30, 60, and 90 for everyone in stretch placement. Expect movement on the skills named in their plans; flat lines trigger more support or role correction. For the new structure as a whole, read coverage columns: did the specialist unit gain a second person at Level 3+ on security? If not, escalate hire or borrow — the restructure is not finished when the org chart publishes.

Link stretch cohorts to onboarding discipline even for tenured internal movers — they are new to the role even when not new to the company. The same ramp logic applies: foundations first, evidence at checkpoints, re-score against required levels.

Which site tools support restructures?

• Before you restructure (overview)
• Upleashed 0–5 methodology
• Descriptor generator
• Excel Skills Matrix Template
• Succession planning
• Workforce capacity planning

How does this guide connect to the rest of the site?

The printable skills-matrix-in-a-restructure.pdf is built for facilitation; use this page when you need live links, extra examples, and site tools in context.

Anchor ratings to the methodology pillar, then generate level wording with the descriptor generator before your first calibration.

Link each matrix review to a decision log (training booked, hire briefed, project staffed) so the grid drives action.

What if the new structure changes again within a year?

Version the target design and keep historical placement records. People may move twice — score against the role they hold today, not the one they left. Frequent change is why dated cells matter more than perfect org charts.

Frequently asked questions

How do I use a skills matrix in a restructure?

Score current capability, define each new role by skills and required levels, map people to matches and stretches, redeploy where possible, form fair pools where headcount must reduce, and plan for gaps the new design exposes before go-live.

Why not rely on the org chart?

The chart shows reporting lines, not who can perform in new roles. Capability data prevents placing strong people where they cannot succeed and losing skills you still need.

How is redeployment different from normal hiring?

Redeployment uses people already in the organisation whose scores show fit or short stretch — faster and cheaper than exit and external hire for the same capability.

Can a matrix replace HR and legal process?

No. It supplies evidence for defensible decisions; consultation, selection rules, and local law still govern.

What is a stretch move?

Someone one level below the new role's requirement on a key skill, placed with a development plan — realistic when support exists, risky when treated as a full match.

How often should we update scores during a restructure?

Baseline before design finalises; refresh at placement and again 90 days after move — roles and people change under stress.