The Second Customer.

The Geschäftsführer of a mid-sized German hospital receives approximately forty-eight workforce metrics per quarter. Turnover rate, overtime by group, sick leave by department, engagement survey composite, time-to-hire, Pflegeschlüssel compliance, on-call burden, training completion. Forty-eight numbers. All accurate. None sufficient.

The aggregate turnover figure itself — approximately one in six hospital employees changes job annually in Germany [1] — sits in the quarterly deck as a single percentage. The actual story sits somewhere else entirely.

In the three weeks before each quarterly executive committee, the controlling team compiles these numbers into a deck. The deck is reviewed. Decisions are taken — or deferred. And somewhere on a ward two floors below, a senior nurse named Andrea has been telling her colleagues for six months that the department is coming apart. Four of the five experienced staff have mentally decided to leave. The scheduling system introduced in January is silently destroying the goodwill a decade of leadership carefully built. A specific Oberarzt is driving people out and nobody above his level has been told.

None of this is in the forty-eight metrics. All of it is knowable to anyone who walks the ward and asks.

This is the diagnostic problem at the foundation of every hospital’s workforce layer: your staff know things your controlling reports do not, and most hospitals have never built the discipline that closes the gap between the two signal streams.

Staff are your second customer.

Most hospitals operate as if patients are the only customer. The logic is obvious — patients pay, patients are the reason the organisation exists, patients are who regulators protect. Every hospital I have worked in, inspected, or consulted to, has a Patient Experience function with dashboards, survey programmes, and board-level visibility.

Very few hospitals operate as if staff are a second customer.

The distinction matters because it changes the question you ask. A customer is someone whose continued engagement is a choice they make daily, whose trust you have to earn rather than assume, and whose defection is a signal about something you did — not something wrong with them. When you treat staff as employees, their departure becomes an HR event, catalogued in the turnover statistic. When you treat them as a second customer, their departure becomes a diagnostic finding.

The shift in language sounds cosmetic. In practice it is structural. A hospital that has understood its staff as a second customer has typically done three things the ordinary hospital has not. It has made frontline staff feedback a signal at least as important as controlling reports, not a complement to them. It has built the leadership infrastructure to hear that signal — not as an engagement survey delivered annually, but as routine presence and routine listening. And it has been prepared to act on what it hears even when the action contradicts what the quarterly numbers seem to recommend.

Those three things are rare. Together they are rarer still.

The gap between signal streams is itself a metric.

Every hospital with staff has two parallel signal streams running at all times. The first is the one leadership sees — the aggregated, time-lagged, upward-flowing stream that produces controlling reports. The second is the one the wards produce — continuous, specific, and almost entirely invisible to leadership unless it is deliberately surfaced.

The diagnostic question is not whether both streams exist. They always do. The question is how large the gap between them is, and whether leadership has installed any mechanism to measure that gap.

In a well-run hospital, the two streams are largely congruent. What the controlling report shows is what the wards would tell you. When staff name problems, leadership recognises them from the data. When leadership asks why a number moved, staff can explain it in the language of their daily reality. The streams reinforce each other.

In a badly-run hospital, the streams diverge. The controlling report shows a stable turnover rate; the wards know four key people have already decided to leave and the number will spike in six months. The engagement survey shows a 3.8/5 composite; the wards know the score is carried by two high-engagement departments while three others are in structural collapse. The Pflegeschlüssel compliance number shows green; the wards know compliance is being gamed by rotating the same few staff across understaffed departments on paper.

The size of the gap is the single most diagnostic marker of the People layer. Not the turnover rate. Not the engagement score. The gap.

The three numbers most CEOs cannot recite.

The Five-Layer Diagnostic I have published elsewhere opens its ANCHOR section with a specific question: can your leadership team state, from memory, your hospital’s aggregate annual turnover rate, average overtime hours by group, and sick leave rate — disaggregated by department — without consulting a report?

The question is not rhetorical. I have asked variants of it in settings where hiding a poor answer was difficult — scoping calls, board meetings, strategic retreats. The pattern is recognisable across the senior leadership teams I have worked with.

A small minority of leadership teams can state the aggregate figures from memory and name the three departments with the highest readings on each of the three metrics and link those names to specific, recent operational events. That minority is the population whose ANCHOR layer is actually being monitored. The rest have workforce metrics that sit in reports, not in working memory — and the wards know it.

Walking the wards is not optional. It is the measurement instrument.

The closer answer is that most controlling reports were never designed to detect the problems that cost hospitals their senior staff. They were designed to detect problems that appear at the aggregate level once enough people have already acted. A controlling report can detect a department that has already lost three nurses. It cannot detect a department where four nurses have mentally decided to leave and have not yet handed in notice.

The instrument that detects the second situation is one specific leadership behaviour. You walk the wards. Unannounced. Regularly. Without an agenda. And you talk to people.

This sounds almost insultingly simple. In practice, hospital leaders resist it on roughly four grounds. The first is time — walking the wards is four to six hours per week that cannot be reallocated elsewhere. The second is discomfort — conversations with frontline staff about real problems are harder than meetings about numbers, because numbers do not push back and people do. The third is a professional identity issue — walking the wards feels like the work of a supervisor, not a CEO. The fourth is an implicit belief that the controlling report is the organisation’s canonical source of truth, which ward-floor observation is expected to confirm rather than correct.

All four objections are symptoms of the same underlying error: treating staff as employees rather than as a second customer.

A customer-intelligence function is one that measures its customers through direct contact, not through proxies. A hospital that has understood staff as a second customer treats ward walks the way an airline treats customer interviews — routine, structured, taken seriously, and reported on at the same leadership cadence as the financial numbers. The Geschäftsführer has a weekly walk schedule. The Medical Director has one. The Pflegedirektorin has one. What they hear is written down — not in a surveillance sense, but in a diagnostic sense — and fed into the same meeting where the controlling report is reviewed.

The gap between the two signal streams becomes itself the primary metric.

One case, specifically, where this held.

The ward-walking discipline described above is specialty-agnostic. It applies in the same form to surgery, internal medicine, intensive care, oncology, and geriatrics. It applies to community hospitals and university hospitals. The mechanism — that ward staff hold operational signal that controlling reports cannot capture in time — is a feature of hospital organisations in general, not of any one specialty.

What follows is one case in which the discipline was installed deliberately and the results were sustained and measurable.

Between 2019 and 2025, the geriatric department at Main-Kinzig-Kliniken Schlüchtern underwent a structural transition from a loss-making unit into the highest-margin department in its hospital group across three consecutive fiscal years. The operational elements of that transition are documented in the formal research programme conducted under academic guidance of Prof. Dr. Rainer Sibbel at Frankfurt School of Finance & Management [2,3].

What is less covered in the published operational and financial work — because it is harder to quantify than LOS or CMI — is the ANCHOR layer discipline that preceded it.

In the first eighteen months of the transition, the department lead spent approximately four hours per week walking the wards. Not observing. Listening. The conversations were specific: what is frustrating you this week; what did you try to fix that didn’t hold; what would you do differently if you were running the department. Nothing was promised. Nothing was immediately acted on. The point was presence, and the accumulation of a parallel signal stream that existed alongside the departmental controlling report.

The gap, during that period, was substantial. The controlling report showed stable turnover; the wards revealed three specific individuals — one Oberarzt, one Stationsleitung, one long-serving Pflegekraft — whose continued presence was holding the department together against structural drift. Controlling data showed average overtime; the wards revealed that the majority of overtime was concentrated in a single ward where the scheduling model was broken in a specific, fixable way. Engagement scores were middle-of-the-pack at department level; the wards revealed that the composite was carried by one ward where morale was genuinely high while two others were in quiet collapse.

The structural turnaround began with those observations. Not with restructuring. Not with new hires. Not with investment. With listening, and with acting selectively on what was heard. The same pattern has been observed in subsequent engagements across different hospital contexts; Schlüchtern is the longest-running and most formally documented case.

What the ward-walking discipline actually looks like.

The discipline is not complicated. What makes it rare is consistency, not sophistication. Four components distinguish leadership teams that have installed it from leadership teams that have not.

First — the walk is routine, not performative. It happens on the same days each week. Staff stop reacting to it as an event. They begin answering questions honestly because honest answers cost them less than performative ones. An unannounced walk that happens every Tuesday is far more diagnostic than an ambitious monthly tour that everyone knows is coming.

Second — the walk is structured around listening, not delivery. You are not there to communicate strategy, reinforce culture, or recognise people’s work — all of which are defensible leadership behaviours that do not happen to be ward-walking. You are there to collect signal. The three questions that consistently produce useful output are: what is slowing you down this week, what have you given up trying to fix, and what would you change if you were running this department tomorrow.

Third — what you hear is written down. Not as surveillance. As a parallel signal stream that sits next to the controlling report at the same cadence. When the two streams diverge, the divergence itself becomes the agenda for the next operational review.

Fourth — selective action follows, and is visible. The failure mode of ward-walking programmes is hearing a hundred things and acting on none of them. Credibility is destroyed faster by visible inaction than by absent leadership. The discipline is to identify the two or three highest-signal items from each week, to act publicly on them within the next four weeks, and to close the loop with the staff who surfaced them. Everything else is noted, deferred, and not promised.

Four components. None of them requires a consultant. All of them require leadership time that most Geschäftsführer have not decided is worth protecting.

The diagnostic implication.

If you have reached this point and you are wondering whether your hospital’s ANCHOR layer is holding, there is a specific and uncomfortable test. It is the first question in the Five-Layer Diagnostic, and it is asked the way it is asked for a reason.

Can your leadership team — you and the two or three people who sit closest to you in the operational hierarchy — recite, from memory, your hospital’s aggregate annual turnover rate, average overtime hours by group, and sick leave rate, disaggregated by department? Not pull the report. Recite.

If the answer is no, the finding is not that you need better reports. The finding is that you and the wards are running two signal streams that have never been brought into the same room, and the gap between them is quietly costing you senior staff, operational continuity, and the durable goodwill that eventually becomes your margin advantage.

Staff are your second customer. Most hospitals have not made the shift to understanding them that way. The ones that have, operate at a different level — and the financial results, at a hospital like Schlüchtern, follow from that shift rather than producing it.

Walking the wards is not optional. It is the measurement instrument.