Start with the problem, not the new structure.
Reorganizations often begin with a solution ("we should move to business units") rather than a diagnosis. Write down, in one or two sentences, what the structure is preventing: enterprise deals stalling between sales and product, duplicate work across two regions, managers with too many direct reports to coach anyone, or decisions that always escalate to the CEO. Then ask what the new structure would let you do that you cannot do today. In the same McKinsey survey, two of the most frequently cited objectives of redesigns were improving decision making (40%) and cutting costs (39%). Both are legitimate, but they lead to very different designs, so be explicit about which one you are optimizing for.
Rule out cheaper fixes first.
Many problems that look structural are really decision, priority or people problems. Before you change reporting lines, check this list: Unclear decision rights: if two teams argue about who decides, define the owner and an escalation path. Conflicting priorities: if teams pull in different directions, the leadership team has not made a trade-off yet. One role, not the structure: if the issue disappears when you mentally swap one person, it is a hiring or performance question. A missing process: a weekly decision forum, a deal desk or a shared planning cycle can fix handoffs without moving anyone. Structure is the right lever when the problem persists regardless of who holds the roles and is built into how teams report, are measured and are funded.
Use five questions before you commit.
A short framework keeps the debate on evidence rather than on whose team gets bigger:
- Problem: which outcome is the structure blocking, and what evidence shows it?
- Options: compare at least three paths, for example a targeted change (move one team or change decision rights), a partial redesign of one area, and a full reorganization, plus doing nothing.
- Duration and distraction: plan for longer than you expect. McKinsey's respondents said nearly 30% of fully implemented redesigns lasted more than 12 months from setting objectives to completing all changes.
- Reversibility: identify the parts that are hard to undo, such as senior role changes, redundancies, customer-facing team changes and agreements with employee representatives. Our guide to reversible vs. irreversible decisions explains how to tell them apart.
- Success measures: agree on two or three metrics and a review date before the announcement, not after.
More on Enterprise Transformation — every week.
Join 2,400+ leaders getting the decision intelligence briefing.
Know the base rate for large-scale change.
Restructuring is often part of a broader transformation, and those have a poor track record too. A 2024 BCG study of transformations at nearly 2,000 public companies between 2001 and 2022 found that only 26% created value in both the short and the long term. These figures come from large and listed firms, and the McKinsey numbers are self-reported and more than a decade old, so treat them as a warning rather than a forecast for a 150-person company. Smaller firms can move faster with fewer layers, but they also depend more on a handful of key people, so losing one during a reorg hurts more. A pre-mortem is a useful way to surface those risks before you decide.
Hypothetical example: functions or business units?
Imagine a 180-person B2B software company organized by function. Enterprise deals keep stalling because sales, solutions engineering and product disagree about custom requests, and the CEO is considering a switch to two business units. Working through the five questions, the leadership team finds that the real problem is that no one owns the decision on custom requests; the functional structure works well for everything else. They compare three options: a business-unit reorganization that would take an estimated nine to twelve months, a deal desk with a named owner and a weekly decision slot, or no change. They pilot the deal desk for one quarter and track time from request to answer and the enterprise win rate. If deals still stall, the business-unit option comes back to the table with much better evidence. This example is illustrative, not a client case.
Plan the legal and people side early.
If the restructuring affects jobs, locations or ways of working, employment law and co-determination rules may apply. In Germany, for example, a major restructuring in a company with a works council (Betriebsrat) can count as a Betriebsänderung that requires information and consultation before you implement it. Rules differ by country and company size, so involve employment-law counsel before the decision is announced; this article is not legal advice. Equally important is the explanation: people accept change more readily when they understand which problem it solves and how you will know whether it worked.
Get the decision challenged before the org chart is drawn.
The most expensive reorgs are the ones decided in one leadership offsite and defended afterwards. If you want an independent challenge first, NeuroAgent's AI council can test whether the problem is truly structural, compare targeted options with a full reorganization and check the result against your constraints on budget, headcount and timeline; a Decision Sprint turns that into a decision-ready brief for your leadership team or board, and you can try the council free without an account.
Research context
BCG found that only 26% of corporate transformations created value in both the short and long term over the two decades studied.
Boston Consulting Group: Five Truths (and a Lie) About Corporate Transformation (2024)This independently published statistic provides context for the topic; it does not substantiate every claim in this essay.
One decision framework, every week.
The week's best articles on making and executing high-stakes decisions, backed by sourced research.