
Here’s a question I think people should be able to answer. And by ‘people’, I mean any leader in a change process.
What does a year of restructuring cost the people who stay, and does the cost stop when the process does?
I note the restructure of FENZ is in the papers today. Regional councils have to restructure, but we’re not sure how. Or exactly when? Maybe you’ve been restructured. I’ve both led restructures, and been restructured.
Let’s go back a bit. A 2016 systematic review pulled together 39 longitudinal studies published between 2000 and 2012. Its conclusion was pretty blunt and, I think you’ll see, unsurprising. Restructuring predominantly harms employee wellbeing. The review couldn’t establish whether the effects differ between restructuring with staff reductions and restructuring without them, because negative findings turned up in both. Among the studies investigating the impact on wellbeing, 18 of the 23 with staff reductions and eight of the nine without them reported negative findings. An implication of that is not to treat the absence of redundancies as a reason to skip the question.
It gets more uncomfortable though.
Among the studies that measured people before the change was even announced and then followed them through more than two time points, most provided no support for a recovery in wellbeing. The before-and-after windows in those studies ran from two to eight years. A couple of studies did find recovery, and in some cases people still hadn’t returned to where they started.
In other words, the end of a restructure is not reliably the end of its effects, and a leader who treats the announcement as the closing of the file is making an assumption the evidence doesn’t support. Again, I don’t think this is surprising. I also think leaders would, at least academically, acknowledge the truth of it without needing to read the research.
For transparency
Now the part where it gets more complicated, so you can see I’m not just cherry-picking data,
A 2025 French study followed 15,185 employees who all had good psychological wellbeing at the start and asked them whether their work environment had changed strongly in the previous year. People who’d been through any big change at work developed poor wellbeing about a third more often. That part was clear.
Then the researchers split the changes up by type, and the picture broke apart. Restructuring and relocation came out slightly worse than no change at all, and so did a redundancy plan, but in both cases the results were too uncertain to say whether there was any real difference. The types that did show a clear effect were a change in technology, a change in how the work was organised, and a buyout or a new management team.
A 2025 Swedish study took a different route. It tracked 17,599 observations from 10,358 workers, measured downsizing from national registers rather than by asking anyone, and counted medically certified sick leave of more than 14 days for common mental disorders. Downsizing made no difference to that measure. People in workplaces that had downsized took long-term sick leave for mental health problems at the same rate as everyone else. The study was large enough to rule out a big difference, not a small one.
Before you think this resolves the issue, here’s another study using 2011 to 2015 data from 19,173 employees in one Norwegian health trust, with downsizing taken from the register. For sickness absence it compared employees with themselves over time, during downsizing and stable periods. The probability of a short spell of sick leave was 4.9 percentage points higher in the quarter after their unit downsized than in their stable periods. Long-term sick leave showed no relationship.
The way I see it is that the results depend on what was changed, who stayed in the study, and what the researchers chose to measure. The 2016 review identifies limits, noting fewer than half of its studies included a comparison of participants who weren’t restructured. The newer work is mixed rather than reassuring, so I’m not reading more into it.
So what recurs across all of it?
In the 2016 review, only three process factors were reported by more than one of the 39 studies. One, communication and providing information about the change, in four studies. Two, procedural justice, in two studies. Three, training related to the change, in two studies. Each related positively to wellbeing. The review is careful that these were tested on their own, with no analysis of how they work, so treat them as things that travel together rather than as proven levers. But you see where I’m going, I hope.
And they keep turning up.
In the French data, among people who had been through a change, those who said they hadn’t been given adequate information developed poor wellbeing about 60% more often than those who had. Those who said they’d had no influence, about 80% more often than those who had some. And the people missing information, consultation and influence all three developed it more than twice as often as those who got all three.
Crucially, those gaps are the norm rather than the exception. Among people who’d been through a change, four in ten weren’t told what was happening, six in ten weren’t asked for their view, and eight in ten felt they’d had no say. Fewer than one in five got all three. These are what employees reported, not a count of employers.
The pointiest bit version is the Norwegian piece. Out of every hundred people whose employer hadn’t downsized, about four developed mental distress. Out of every hundred who’d been through a downsizing they thought was handled well, about five. Out of every hundred who thought it was handled badly, about sixteen. Some people in every group developed distress, and most people in every group did not.
That’s useful to note for next time. Whether your wishes and input were taken into account in planning and carrying out the change. Whether you had the opportunity to speak with your immediate leader about what it meant for you. Whether you got the training you needed for new tasks or a new role. Five-point agreement scales, so it asked for judgements rather than yes-or-no records.
Again, a few limits, for transparency. These are associations in observational studies, not trials. In the Norwegian data the process rating and the symptoms were collected in the same survey, so someone already struggling may rate the process more harshly, and excluding people who were distressed at the outset helps without settling it. And the questions below identify gaps in how a change was handled. They don’t tell you whether anyone has recovered, so keep whatever wellbeing checks and support routes you already have running after the process ends.
Here’s what I think is practical
Those three questions ask what the organisation did, not how anyone is feeling, so nobody has to notice their own distress, name it correctly, or put it on a form for HR. They don’t need a new exercise either. The 2016 review’s own recommendation is that restructuring variables can be included in existing company surveys. If you already run one, these go in it.
My addition would be to include the three adapted questions to that survey, for anyone who experienced organisational change in the past year rather than only those whose job title changed, because the manager, the team and the work can all change around someone whose role stayed the same. Agree before it goes out who can see individual answers and how small groups will be reported, because calling something an aggregate doesn’t protect a team of four when eveeryone is likely identifiable.
Then set the action, the owner and the review date while the change team can still own them. When you report back, say what was actually done and ask the people affected whether the gap closed.
Photo by Julia Taubitz on Unsplash
