Workplace adjustments are one of the most effective tools an employer has for keeping people in work. Almost no organisation can say how many of its people are on one today, where they are, when they end, or what they cost. That gap is about to become visible.
Five questions, and why they go unanswered
Ask an employer how many of their people are currently working under a workplace adjustment. Not how many were agreed last year. How many are in place this morning.
Then ask where they are. Which sites, which teams, which managers.
Then ask when each one ends and when it is next reviewed.
Then ask how an adjustment starts, and what triggers it.
Then ask how that joins up with everything else already running: the return-to-work interview, the occupational health report, the request an employee made to their manager in a corridor, the flexible working request sitting in an inbox, the case that has been open since the summer.
Very few employers can answer the first question, and almost none can answer all five. This is not because the information is sensitive or difficult. It is because it was never collected in one place in the first place.
There is a cost question underneath it
The visibility problem carries a financial one that rarely gets asked out loud.
Of the people currently on an adjustment, how many are on reduced hours, a reduced role or different duties, and still being paid in full? For how long has that been the case? What was agreed at the time, by whom, and on what basis?
For most employers this is unanswerable, which means the cost is real, continuing and unquantified. It does not appear as a line in any budget. It is absorbed into payroll and never surfaces as a decision anyone made.
Adjustments are not run as a process anywhere
The reason all of this is invisible is structural rather than administrative.
Adjustments are not a process in most organisations. They are a series of individual conversations, bolted onto the side of four or five other processes that do have a shape.
One is agreed during a return-to-work interview and lives in that form, wherever that form lives. Another emerges from an occupational health report and lives in a PDF attached to an email. Another was a quiet arrangement between an employee and a manager two years ago that both sides now treat as permanent. Another was requested and never formally answered at all.
None of those records can see each other. None of them roll up into a number. And because no single function owns the whole thing, three predictable failures follow.
One: agreed is not the same as in place
There is a date the employee raised it. Then there is a gap.
The manager is not certain what they are permitted to agree. The occupational health report recommends consideration of amended duties, which is not something anyone can action. Something needs to be purchased, or a rota needs to change, and that has no owner. The employee asks once, gets no clear answer, and stops asking.
Nothing was refused. Nothing happened either.
Then the absence the adjustment existed to prevent begins again, and the second episode is usually longer than the first. The organisation has absorbed all of the cost of the problem and none of the benefit of the solution.
Very few employers can produce the two dates that matter here: when an adjustment was agreed, and when it actually became operational. Without those, the gap cannot be measured, so it is never managed.
Two: nothing ends and nothing is reviewed
An adjustment with no end date is permanent by accident rather than by judgement.
That cuts both ways, and both ways are a failure. Some people could have come off an adjustment months ago and would have preferred to, but nobody revisited it. Others needed more support than they were given, and were never asked again.
A review date that depends on a manager remembering is not a review date. It is a hope.
Three: adjustments do not survive a handover
This is the failure that quietly undoes the work that did get done.
The manager changes. A restructure, a promotion, somebody leaves. The arrangement was verbal, agreed between two people, and one of them has gone. The new manager inherits a team and no history.
So the employee either re-explains a health condition to a stranger, or decides it is not worth the conversation and goes back to doing the job the hard way.
The organisation agreed that adjustment. It paid for it. It still carries the legal duty attached to it. And it lost it in a handover that nobody would describe as a failure at the time.
The legal position makes the record the defence
Under the Equality Act 2010, employers have a duty to make reasonable adjustments for disabled employees. Where that duty is tested, it is tested on documents: what was requested, what was considered, what was agreed, what was refused and on what grounds, and whether anything was reviewed.
An employer whose adjustments live in notebooks, inboxes and individual memory cannot produce that record. It has to be reconstructed after the fact, usually eighteen months later, usually by people who were not involved at the time.
The expectation is also moving. The Keep Britain Working review has pushed employers toward acting early and being able to evidence that they did, with named actions, owners and review points rather than general statements of intent. The questions that follow from that are date questions. An employer either holds the dates or it does not.
What running adjustments as a process looks like
It is not complicated, and it is not expensive.
Every adjustment has a start date, a named owner, and a decision with a reason attached, including when the answer is no. A refusal with a recorded rationale is a defensible position. A request that was never answered is not.
Every adjustment has a date it actually became operational, which is a different field from the date it was agreed, because the distance between those two dates is where the cost sits.
Every adjustment has a review date that arrives whether anyone remembers it or not.
Every route in lands in the same place. A return-to-work interview, an occupational health recommendation, a request made directly to a manager and a flexible working request should all create the same record, rather than four records in four systems that never meet.
And the record travels with the employee rather than with the manager. When the manager changes, the next one picks it up on day one, and the employee does not have to disclose anything twice.
What that gives you
The five questions above acquire answers. A number, broken down by site, by team and by manager, with review dates attached. A clear view of who is on reduced hours or duties and still on full pay, and for how long.
Beyond the reporting, the operational returns are immediate. Absences are shorter, because the intervention that gets someone back to work happens in days rather than months. Fewer cases escalate, because an employee whose adjustment is honoured does not need to raise a grievance to be heard. Turnover falls in exactly the teams where replacing people is hardest. And the evidence trail exists already, built as the work happened, rather than assembled under pressure after a claim lands.
Where to start
Pick three people in your organisation who are on an adjustment today.
Find out when each one was requested, when it was agreed, when it actually became real, when it is next reviewed, and who would know if it had quietly stopped.
If that takes more than ten minutes, you have your answer.