Ask most Operations Directors who handles the difficult people cases in their part of the business and the answer comes back quickly. HR does. HR advises the managers, HR owns the process, HR keeps it all on the rails.
It is an entirely reasonable belief. There are policies. There is an HR function. There is very probably a service desk with a number to ring. When something difficult lands on a manager — a grievance, a long-term absence, an employee hitting an attendance trigger — they contact HR and HR tells them what to do.
That is the design. It is frequently not the reality, and the gap between the two sits almost entirely out of the Operations Director’s line of sight.
Start with the word
The problem begins with the word itself, which is doing far more work than anyone acknowledges.
In a large employer, “HR” describes at least three quite different jobs.
There is an HR Director or CPO operating at board or senior leadership level. They know the business, they understand commercial pressure, and they will tell a fellow Director something they do not want to hear.
There is a business partner attached to a region or a division. At their best, they know the managers personally, they have walked the site, they understand what a Monday morning actually looks like there, and they have enough relationship capital to be direct.
And there is a shared service advisor working a queue. They have not met the manager. They have not seen the operation. They have a policy document, a case note and a ticket number.
Three different roles, three radically different levels of context, and one label covering all of them. The manager with a problem gets whichever one picks up.
Distance degrades advice
The pattern is consistent enough to state as a rule: the further HR sits from the manager, the vaguer the advice becomes.
This is not a criticism of the people involved. An advisor who has never met the manager, never visited the site and knows nothing about the operational context has far less to work with. Faced with an ambiguous case and no way to judge the surrounding facts, the rational response is caution. Caution sounds like this:
“It isn’t clear.”
“That’s too risky.”
“Let’s close this one and do better next time.”
None of those is an answer a manager can act on. And because the response depends on which advisor happens to pick up the case, the same question asked a fortnight later can come back differently. In our experience it is not unusual for the advice to be a personal interpretation of the employer’s own written policy rather than the policy itself.
Alongside that, the operating focus shifts. A function measured on queue performance optimises for queue performance. The question quietly moves from what does this business need here to how many tickets did we close this week.
Why that is a risk question, not a service question
This is the point at which it stops being an HR service-quality issue and becomes something a board should care about.
If the answer a manager receives depends on who answered the phone, then policy is not being applied consistently. That is not a soft observation about customer service. Inconsistent application of policy is the precise thing that gets examined when a case is challenged, and it is the difference between a defensible decision and an expensive one.
It shows up in ways that are easy to misread as case complexity. One large retailer told us their target for cases requiring an external ER advisor was five per cent of caseload; the actual figure had crept to three times that. The cases had not become more complex. Managers had stopped attempting them. Another organisation described case documentation as constantly protracted — assembling the evidence for a single complex case was taking five to ten days of manual work, not because the case demanded it but because nothing was structured well enough to do it faster.
Both of those look like symptoms of difficult casework. Both are symptoms of managers who cannot get a usable answer.
What the Operations Director can see
Very little, is the honest answer.
What reaches an operational leader is days lost. Sometimes a case count. Neither of those says anything at all about whether a single case was actually managed — whether the call was made at the right time, whether the return-to-work conversation happened and meant anything, whether the manager was told what to do next in terms they could act on.
So the picture available to Operations is a number that looks broadly stable, alongside a private suspicion that managers are not really engaging with absence and case work, and an observation that the problems present last quarter are still present this quarter.
The cost surfaces later, and in someone else’s report. Lower productivity in the affected areas. Higher agency and overtime spend. On contracted operations, thinner contract profitability. By the time it is visible in those terms, the trail back to a case that stalled eighteen months ago because nobody could get a straight answer is long cold.
The measurement gap
Underneath all of it is a single absence in the data.
Organisations measure tickets raised, time to close, cases open, cases resolved. Throughput. Nobody measures whether the advice given was correct, whether the manager was able to act on it, or whether the case genuinely resolved rather than simply closing.
Because that measure does not exist, the numbers look healthy while the cases quietly do not. And no amount of scrutiny applied to the existing reporting will surface it, because the existing reporting was never designed to answer the question.
This is not an argument against HR
It is worth being clear about that, and it is worth noting where the sharpest version of this argument comes from.
Raising this publicly recently, the most pointed response came not from an operator or a supplier but from a senior HR leader. Her position was that if HR is transactional and does not understand the business, the operation and the way its leaders work, she would not place much value on its advice either — and that her real frustration is with people carrying the title business partner without the business understanding the title implies.
That is the argument being made from inside the function. The people in a service centre are not the problem. Those models were designed to reduce cost and were optimised for volume, and they largely achieve what they were built to achieve. Advice quality was simply never one of the measured outputs.
What to do about it
Three questions, and none of them requires a project.
Ask your managers who advised them on their last difficult case. Not which department — which person, and what they were told.
Ask whether they could act on it. This is the question that produces the useful answer. “I rang HR” is not the same as “I was told what to do.”
Then look at your own data and ask why none of this was visible. If the reporting you receive could not have told you any of the above, that is the finding.
Accountability for the sites, the managers and the risk sits with Operations. Accountability for the reporting sits with HR. Assuming that HR has both you and your managers covered is one of the more expensive assumptions available to a large employer, and it is rarely tested until something goes wrong.
