WHY RESTRUCTURES COME APART IN THE SECOND MONTH

Restructures rarely fail at the announcement. They come apart weeks later, for four reasons that are visible while the design is still on paper.

Emmanuel Duncan · 7 September 2026 · 6 min read

Most restructures I have watched close up follow the same shape. The design work is careful, the announcement is handled properly, and the first fortnight goes better than anyone expected. People are relieved that the uncertainty is over. Managers report that the team took it well.

Then somewhere in the second month the energy leaks out. Work that used to move starts sitting. Two managers each believe the other owns a decision. Someone senior begins holding informal meetings that duplicate the formal ones. By the end of the quarter the organisation has a new chart and largely the old behaviour.

The causes are visible in advance, and none of them are about communication in the way that word is usually meant.

The First Cause: The Chart Moved and the Work Did Not

A restructure changes reporting lines. It rarely changes the actual sequence of steps by which work gets done, because that sequence lives in habits, in shared files, in a WhatsApp group, and in the memory of four or five people who know how to get things unstuck.

In the first weeks those habits carry the organisation through. People route work the way they always did, using relationships that predate the change. That is why the early period looks calm. The strain appears when those routes start failing, usually when someone who was a node in the old network moves, leaves, or decides the new arrangement means it is no longer their job.

The preventable version of this is to map the three or four processes that matter most before the design is finalised, at the level of who does what after what, and check the proposed structure against them. Where a process now crosses a boundary it did not cross before, that handover needs a named owner and an agreed trigger, written down.

The Second Cause: Accountability Was Distributed and Never Landed

Restructure documents describe responsibilities. They are much weaker on decision rights, which is a different thing.

Responsibility says who does the work. Decision rights say who can approve, who must be consulted, and who can say no. When a design leaves those unstated, the gap does not show immediately, because the first weeks are full of transitional decisions that everyone agrees to escalate. It shows when the first genuinely contested decision arrives, and two people both believe it is theirs.

The test I use is deliberately concrete. Take six decisions the organisation actually made last quarter, real ones with names and amounts attached, and ask who would make each of them under the new structure. If any of the six produces hesitation or disagreement in the room, that is the decision that will stall in week six.

The Third Cause: Nobody Was Given Time to Stop Doing the Old Job

People appointed to new roles almost always keep the old one for a period, because there is nobody else to hold it and because the work does not pause for the transition.

Where this is handled explicitly, with a named end date and a named successor, it works. Where it is left implicit, the new role never starts. The person spends their days on the familiar work, which is urgent and which they are good at, and the new responsibilities become the thing they will get to when things settle.

Two months on, the structure exists on paper and the organisation is running the previous one with new titles.

The remedy is unglamorous. For each person whose role changed, write down what they are stopping, who is picking it up, and the date. Then check it at the end of the month rather than at the end of the quarter.

The Fourth Cause: The Informal Structure Was Never Acknowledged

Every organisation has people whose influence exceeds their position. The long serving officer everyone consults before committing to anything. The administrator who knows where the records are. The supervisor whose opinion sets the tone for a whole shift.

A restructure that ignores these people does not remove their influence. It removes their formal standing while leaving the influence intact, which is the worst combination, because they now shape opinion without any responsibility for the outcome.

Identifying them is not difficult. Ask a handful of people in different units who they would call to get something unstuck, and the same names appear. Bringing them in early, with a real role in the design rather than a briefing after the fact, changes the second month more than any amount of announcement.

What to Watch for in Weeks Five to Eight

Four signals, all of them observable without a survey.

Decisions taking longer than they did before the change. Ask two or three people to name a decision they are currently waiting on, and find out where it is sitting.

Meetings multiplying. New coordination meetings appearing on calendars are a sign that the structure is not carrying coordination on its own, and each one is a small piece of evidence about which boundary is not working.

Escalation to the person who used to hold the role. If people still route questions to the previous holder rather than the current one, the handover is unfinished regardless of what the chart says.

Quiet exits among people who were not affected by the change. Regretted departures from teams that were left alone are worth reading closely, because they often reflect a judgement about where the organisation is heading rather than about the restructure itself.

The Baseline You Will Wish You Had

Almost nobody measures anything before a restructure, and so almost nobody can say afterwards whether it worked. The evaluation becomes an argument between people who wanted it and people who did not.

Four measures taken in the month before the change cost very little. Time from request to decision on a common approval. Vacancy count and average time to fill. Regretted exits over the previous twelve months by unit. And one short question on role clarity, put to the affected population, with the same wording you will use again later.

Take them again at three months and at nine. Even a rough baseline changes the quality of the conversation, because it moves it from impressions to something both sides can look at.

Where to Start

If you are designing a restructure now, take the six decisions test into the next design meeting and see how many are clean.

If you are already in the second month and it feels heavier than the first, start with the decisions people are waiting on. List them, find where each is sitting, and name an owner for every one. That exercise takes a day and it usually surfaces the two or three boundaries that were left ambiguous in the design.

Structures do not fail because people resist them. They fail because the work was never rerouted to match.

Emmanuel Duncan is the founder of Aletheia Lunara Consult, a professional education and consulting practice working with organisations across Ghana and beyond on people analytics, workforce research and organisational capability.