How to Plan an Organizational Restructure: Building the Design Case, Consulting People in Time, and Managing the Transition Without Losing Capability

Most Restructures Go Wrong in the Sequence, Not in the Org Chart

Published 2026-10-12 · EuroQuest International

Quick summary

  • Start with the problem, not the chart. A restructure fixes problems caused by structure: duplicated work, unclear decision rights, too many layers. It does not fix weak leadership, broken processes or missing skills.
  • Design roles before choosing people. Agree the design criteria and the roles first. Putting names in boxes too early turns a design exercise into a negotiation about individuals.
  • Consultation is part of the plan, not a step after it. Where job losses are proposed, many jurisdictions require consultation while the proposals can still change, and set minimum periods before anyone is dismissed.
  • Selection has to be fair and documented. Pools, criteria and redeployment options should be written down before anyone is assessed against them.
  • The transition is where value is lost. Knowledge leaves with people, and the staff who remain decide whether the new design works. Plan both explicitly.

Restructures are usually announced as a design decision and remembered as a people event. The new chart is often reasonable. What goes wrong is the order in which things happen: names are attached to boxes before the design is agreed, announcements run ahead of consultation, selection criteria are invented after the shortlist is known, and the people who held the old structure together leave before anyone has written down what they knew. Each of those is a sequencing error, and each is avoidable with a plan that treats design, consultation and transition as one piece of work.

This guide walks through the questions to answer, in order, when planning a restructure: what problem it is meant to solve, how to design the new structure, when legal duties to consult begin, how to select people fairly, how to communicate, and how to manage the transition. It is written for HR business partners, organizational development and transformation leads, and line managers who have been asked to deliver a restructure, and for executives who want to know what a sound plan looks like before they approve one. It is not legal advice; the rules on consultation and dismissal are national, and they differ in ways that matter.

On this page

  1. What problem is the restructure meant to solve?
  2. How do you design the new structure?
  3. When do consultation and legal duties start?
  4. How do you select people fairly?
  5. How should the restructure be communicated?
  6. How do you manage the transition without losing capability?
  7. Where do teams build this capability?
  8. Frequently asked questions
45 days
Minimum consultation period before dismissal in the UK when 100 or more redundancies are proposed, or 30 days for 20 to 99, per GOV.UK guidance on redundancy consultations
60 days
Advance notice that covered US employers, generally those with 100 or more employees, must give before plant closings and mass layoffs, under the federal WARN regulations
20 in 90 days
One of the two thresholds EU member states can use to define collective redundancies, which trigger duties to consult and notify, per the European Commission

What Problem Is the Restructure Meant to Solve?

Write the problem down in one paragraph before anyone draws a chart. A useful problem statement names the symptom, the evidence for it, and why structure is the cause. "Decisions on customer pricing take six weeks because four functions each hold a partial veto" is a structural problem. "We need to be more agile" is not a problem statement at all, and a restructure built on it will be judged on whatever the sponsor happens to notice afterward.

Structure is the right lever for a specific set of problems: work duplicated across units, decision rights that are unclear or held too far from the information, spans of control so narrow that managers supervise rather than lead, too many layers between the people who see a problem and the people who can act on it, or a design built for a strategy the organization no longer has. It is the wrong lever for problems caused by weak leadership, broken processes, poor information or missing skills. Moving those problems into new boxes changes their address, not their nature, and the organization pays the full cost of disruption for no structural gain.

Test the Case Before Committing to It

Three questions sort a real structural case from a reflex. First, could the problem be fixed by changing a process, a decision rule or a role description inside the current structure? If so, try that first; it is faster and cheaper. Second, has the same symptom survived a previous restructure? If it has, the cause probably lies elsewhere. Third, what will be measurably different twelve months after the new structure is in place? If nobody can name it, the case is not ready. Restructuring is one tool within wider organizational change, and the broader people side of change is covered in our guide on how to manage organizational change.

How Do You Design the New Structure?

Start from design criteria, not from options. Criteria are the four to six things the new structure must do well, ranked: for example, a single accountable owner for each customer segment, decisions on standard pricing made within one layer of the front line, specialist expertise pooled rather than spread thin, and costs held within an agreed envelope. Every option is then scored against the same list. Without criteria agreed in advance, design debates become contests between preferences, and the most senior preference usually wins.

Generate at least three options, including one that changes less than the sponsor expects. The smaller option is useful even when it loses, because it forces the case for the larger change to be made explicitly. Most designs organize primarily around one of a small number of axes: function, product, customer or market, geography, or process. Every choice strengthens coordination along its own axis and weakens it along the others, so the design question is which coordination problem matters most, and what mechanisms will handle the ones the structure does not.

Primary axis What it does well What it makes harder
Function Deep expertise, consistent standards, economies of scale Coordination across functions; customer and product ownership is diffuse
Product or service Clear accountability for results, speed within each line Duplicated specialist roles, inconsistent standards between lines
Customer or market Responsiveness to distinct customer needs Shared products and capabilities are harder to manage centrally
Geography Local regulation, language and relationships handled close to the market Global consistency, and the cost of repeating functions in each region
Matrix of two axes Balances two priorities without choosing between them Dual reporting, slower decisions, and heavy reliance on managers' skill

Decision Rights, Spans and Layers

A chart shows who reports to whom. It does not show who decides what, and that is usually where the original problem lives. For each major recurring decision, record who recommends, who decides, who must be consulted and who is informed in the new design. Then check spans and layers: how many direct reports each manager has, and how many levels separate the front line from the top. There is no universal right number. Teams doing similar, stable work can sustain wide spans; teams doing complex, varied or highly regulated work need narrower ones. What matters is that the numbers follow from the work rather than from the grades of the people involved.

Design roles before choosing people. Each role in the new structure needs a purpose, its main accountabilities and the decisions it owns, written before anyone is matched to it. When names go into boxes first, the design bends around individuals, and the organization ends up with a structure shaped by its current incumbents rather than by its work. This is the core of organizational design and structural optimization.

When Do Consultation and Legal Duties Start?

Earlier than most plans assume. Where a restructure proposes job losses, many jurisdictions require the employer to consult while the proposals are still proposals, so that what employees and their representatives say can actually change the outcome. A plan that finalizes the design, selects the people and only then opens consultation has the sequence backward, and in several jurisdictions that is not just poor practice but a legal failure.

The international reference point is the International Labour Organization's Termination of Employment Convention, No. 158. Where terminations for economic, technological, structural or similar reasons are contemplated, Article 13 calls for workers' representatives to be given relevant information in good time and a chance to be consulted on ways to avoid the terminations or reduce their number, and on ways to soften the effect on the people affected, as the ILO Helpdesk guidance on business and employment security sets out. Not every country has ratified the convention, but its logic runs through most national rules.

How Three Jurisdictions Frame It

European Union. The Collective Redundancies Directive requires an employer contemplating collective redundancies to consult workers' representatives in good time, with a view to reaching an agreement. The European Commission's summary says those consultations must at minimum cover "means of avoiding collective redundancies or reducing the number of workers affected," and that the employer must also notify the competent public authority in writing. Member states choose how to define a collective redundancy, using either a sliding scale over 30 days based on establishment size or a single test of at least 20 redundancies over 90 days, and national law then adds its own detail.

United Kingdom. The government's guidance sets minimum collective consultation periods before the first dismissal: 30 days where 20 to 99 redundancies are proposed and 45 days where 100 or more are proposed. Employers must also notify the Redundancy Payments Service using form HR1 within the same time limits. The guidance warns that redundancies made without proper consultation are very likely to be found unfair, and that failing to notify can lead to an unlimited fine.

United States. The federal WARN regulations require covered employers to give 60 calendar days' notice before plant closings and mass layoffs. Coverage generally applies to businesses with 100 or more employees, not counting part-time staff, and some states have their own notice laws, which can be stricter than the federal rule.

A note on the law. The examples above are summaries for orientation, not a complete statement of anyone's obligations, and none of them is legal advice. Thresholds, periods, who counts as a representative and the consequences of getting it wrong vary by country and sometimes by region, and they change. Take advice from an employment lawyer in each jurisdiction affected before any proposal is announced, and treat the minimum periods as floors rather than targets.

What Good Consultation Looks Like

Genuine consultation means sharing the reasons for the proposal, the numbers and roles affected, the proposed selection method and the timetable, then listening to alternatives and responding to them in writing. It is not a presentation followed by questions. Keep a log of every proposal raised, what was considered and why it was accepted or rejected. That record protects the organization, and it often improves the design: the people doing the work frequently spot duplication, risk or a cheaper alternative that the design team missed. Stakeholder engagement in change initiatives covers how to run that dialogue with representatives and affected staff.

How Do You Select People Fairly?

Decide the method before anyone is assessed. Where roles are unchanged or only slightly changed, the incumbent usually moves into the new role directly. Where several people are competing for fewer roles, define a selection pool, meaning the group of employees doing similar work who are at risk together, and justify its boundaries. Then set the criteria, the evidence each will be scored on and who will score, and apply them consistently across the pool.

Good criteria are objective, relevant to the future roles and evidenced: skills the new design needs, documented performance, qualifications, and conduct records where these are reliable. Criteria that cannot be evidenced, or that could indirectly disadvantage people because of a protected characteristic, should be removed before scoring begins. Have at least two people score each candidate, check the results across the pool for patterns before decisions are confirmed, and give individuals a chance to see and challenge their scores.

Look for alternatives at every stage. Voluntary redundancy, redeployment into vacancies elsewhere, reduced hours, retraining for roles the new structure creates and natural turnover can all reduce the number of compulsory exits. Keeping a register of vacancies open to at-risk employees throughout the process is good practice and, in many places, expected. Strategic workforce planning and talent retention is the discipline that keeps those options visible.

How Should the Restructure Be Communicated?

Tell the people directly affected first, in person, and before anyone else hears it. Then brief their managers, then the wider organization, then external audiences, on the same day where possible. Rumor fills every gap in that sequence, and it fills it faster than any announcement. Prepare managers before they brief their teams: what is proposed, what is not yet decided, what the consultation process is, and which questions they should pass on rather than answer.

Be precise about what is a proposal and what is a decision. During consultation, language that presents the outcome as settled undermines the process and the organization's credibility at the same time. Give every affected person a named contact, a written summary of what happens next and the dates that matter to them. Plan for the people who are not affected as well: they will be asking whether they are next, and silence answers that question for them. The channels and sequencing behind this are covered in our explainer on what internal communication is, and leader-level messaging in effective communication strategies for change leaders.

How Do You Manage the Transition Without Losing Capability?

The transition is where most of the value of a restructure is won or lost, and it usually gets the least planning. Three risks need explicit owners.

Knowledge leaving with people. Before anyone in a critical role leaves, list what they know that is not written down: client relationships, workarounds, the history behind decisions, and who to call when something breaks. Schedule handovers with named recipients and check them, rather than assuming a leaving date gives enough time. For roles where the loss would be serious, consider retention arrangements tied to completing the handover.

Work falling between the old and new structures. Map every recurring activity from the old structure to an owner in the new one, including the unglamorous ones nobody put on a chart: month-end reconciliations, regulatory filings, supplier renewals. Gaps usually show up weeks after the go-live date as a missed deadline. A transition period with a few named people responsible for catching dropped activities is cheaper than discovering them that way.

The people who stay. Remaining employees decide whether the new structure works. They have watched colleagues leave, their own roles have often changed, and many will be quietly reconsidering their future. Managers need time with each of them in the first weeks: what has changed in their role, what has not, and what the next six months look like. Resistance at this stage is usually a reasonable response to uncertainty rather than opposition, and it eases when the uncertainty does. Change resistance management and workforce alignment covers how to work with it.

Check Whether It Worked

Go back to the problem statement written at the start. At three, six and twelve months, measure the specific things the restructure was meant to change: decision times, duplicated activity, spans and layers, cost, and the outcome measures that motivated the case. Track unwanted turnover among the people the new design depends on. If the measures have not moved, find out whether the design was wrong or whether the old ways of working have simply reasserted themselves inside the new boxes. The second is more common, and it is a reason to work on processes and management habits rather than to restructure again.

Where Do Teams Build This Capability?

EuroQuest International runs organizational development, design and change programs in Geneva, Madrid, Zurich, Amman and Istanbul. Sessions work from the restructures participants are actually facing where they can share them, so teams leave with a sequenced plan rather than a generic model. The wider program sits under organizational development and change management, and the role that usually leads this work is described in our guide to organizational development manager training. For the full change program around a restructure, see managing change and transformation in organizations.

Frequently Asked Questions

How long does an organizational restructure take?

Longer than the announcement suggests. Diagnosis and design usually take several weeks to a few months, depending on scale and how much data has to be gathered. Where job losses are proposed, consultation then runs for at least the legal minimum in each jurisdiction, which in some countries is 30 to 45 days or more for larger proposals, and selection, appeals and notice periods follow. The transition itself, until the new structure is working as designed, typically takes several months after that. Plan the whole sequence before announcing anything, and resist compressing consultation to meet a date chosen before the legal timetable was checked.

Should employees be told before the design is final?

Where job losses are proposed, often they must be. Many jurisdictions require consultation while proposals can still change, which means affected employees and their representatives hear about the proposal before the design is fixed. Even where the law does not require it, early and honest communication about what is proposed, what is undecided and how people can contribute usually produces a better design and less damage to trust. What should not happen is an announcement that presents a proposal as a final decision, or a long silence while rumors fill the gap.

What is the difference between restructuring and change management?

Restructuring changes the formal design of the organization: units, reporting lines, roles, decision rights and sometimes headcount. Change management is the wider discipline of helping people adopt any significant change, whether structural, technological or procedural. Every restructure needs change management, because a new chart does nothing until people work differently inside it. Many changes need no restructure at all. A useful test is whether the problem lies in how the organization is divided and coordinated; if it lies in processes, skills or leadership, a restructure is unlikely to fix it.

How do you choose selection criteria for a restructure?

Choose criteria that are objective, relevant to the roles in the new structure, and supported by evidence the organization already holds or can gather fairly. Typical examples are skills and capabilities the new roles need, documented performance over a reasonable period, relevant qualifications, and reliable conduct records. Agree the criteria, the scoring scale and the evidence before anyone is scored, apply them to a clearly defined selection pool, use more than one scorer, and check the results for patterns that could indicate unfair or discriminatory effects. Take legal advice on the criteria in each jurisdiction, since what is acceptable varies.

How do you know if a restructure has worked?

Measure against the problem it was meant to solve, using the baseline recorded before the change. If the case was slow decisions, track decision times; if it was duplicated work, check whether the duplication is gone; if it was cost, measure the cost. Review at three, six and twelve months, and track unwanted turnover among people the new design depends on. A restructure that has not moved its own measures within a year usually means the design was wrong or that old ways of working have reasserted themselves inside the new structure, and the response is rarely another restructure.

Plan the next restructure in the right order

EuroQuest International runs practitioner training in organizational design, change leadership, workforce planning and stakeholder engagement across Europe, the Gulf and Asia.

Explore organizational development programs