The Exit Interview Problem. The Nurses You Can Still Keep Never Appear In It.

Last Monday I promised you a closer look at why salary sits so low on your own resignation data. This week: what that ranking is actually built on, what nurses say when someone other than their employer is asking, and why the number that matters most belongs to the nurses who stayed.

By Christina Archer, Founder · Week of September 7, 2026

Your exit interviews tell you that pay is not why nurses leave. That conclusion is resting on a much thinner instrument than almost anyone quoting it realizes.

The number everyone cites was never built to answer the question it is being asked.

▪Your resignation data has five slots and a full year of departures to explain

The ranking comes from a specific question. NSI asks each participating hospital to identify the top five reasons staff RNs voluntarily resigned, selected from a list of twenty. Five slots, one submission, covering an entire year of exits at that facility.

Run that against your own volume. A hospital employing 1,000 RNs at the national 17.6% turnover rate loses roughly 176 nurses in a year. Those 176 departures reach the national dataset as five words chosen by whoever filled out the survey.

That is not a flaw in NSI’s work. It is a survey of hospitals and it is transparent about being one, drawn from 527 hospitals across 40 states covering 262,405 registered nurses in calendar year 2025. The flaw is in how the output gets used. A list built to record what hospitals report is being read as a ranking of what actually drives nurses out the door, and those are not the same measurement.

▪Salary is missing from the top five, and the list it sits on carries no percentages

Here is the full picture, in NSI’s own words. Personal issues, relocation, retirement, career advancement and scheduling conflict sit at the top. Rounding out the top ten: education, salary, commute, working conditions, and workload/staffing ratios.

Two things in that sentence deserve more attention than the ranking does.

The first is that no percentages are published for any of it. A reason placed seventh and a reason placed third may be separated by a rounding error or by a chasm — the report does not say, and anyone quoting a precise rank position is reading a precision the source does not offer. The one movement NSI does flag is retirement, which has climbed steadily and now sits third.

The second is what those bottom five have in common. Education, commute, working conditions and workload are the answers a departing nurse gives when the honest answer is harder to say out loud. Salary is sitting in the company of the safest available explanations, which should make you slower to conclude that pay does not matter and faster to ask what those categories are absorbing.

▪The nurses answering someone else’s survey describe a different job

Set your exit file beside what nurses tell an independent surveyor and the two barely rhyme.

In Nurse.org’s 2026 State of Nursing survey of 2,090 nurses and nursing students, 43% said they were likely to leave the bedside within a year and 23% said they were likely to leave nursing altogether. Staffing was the condition they named as deteriorating: 42% said it had worsened over the year, against 8% who said it had improved.

Now set that 43% against the 17.6% who actually go. Most nurses who tell a survey they are likely to leave are still on your unit twelve months later. They are also invisible to every retention instrument you currently run, since the only nurses your exit process interviews are the ones already gone.

▪Pay does not push nurses out — it decides which ones are still looking

The apparent contradiction between “salary ranks low among resignation reasons” and “nurses are unhappy with pay” dissolves once you notice the two questions point in opposite directions.

Ask why someone left and pay is rarely the trigger. Nurses leave for a move, a family situation, a schedule that stopped working, a retirement date. Ask instead what is keeping nurses at the bedside and the same survey returns an uncomfortable ranking: financial necessity first at 41%, ahead of a convenient work schedule at 32%, commitment to patient care at 28%, workplace relationships at 24%, personal satisfaction at 24%, and a pending retirement or transition at 23%.

Read that list again. The most common reason a nurse is still on your unit is that leaving is not currently affordable. Commitment to patient care ranks third.

That is a retention finding of the first order, and your exit interviews are structurally incapable of surfacing it, since the nurses it describes are precisely the ones who never exit. Pay is not functioning as a departure trigger in this market. It is functioning as a floor — the thing that decides how many of your nurses are actively looking while continuing to show up. When that floor rises somewhere else in your market, the 43% stops being a survey statistic and starts being your vacancy rate.

▪The people you most need to hear from have a reference at stake

There is a second reason the exit file reads the way it does, and it has nothing to do with nursing.

Work Institute, which has run more than a million exit interviews across North America, estimates that 70 to 80 percent of exit interviews are still conducted internally — by the employer, usually by an HR partner who works alongside the manager in question, occasionally by the manager. A nurse three weeks from a new start date, who may need this organization as a reference and may want to come back in five years, is being asked to explain her departure to the organization she is departing.

She will say relocation. She will say career advancement. She will say the commute. Every one of those answers is true enough to survive scrutiny and safe enough to cost her nothing, and every one of them lands in a category on that list of twenty.

The categories are not lying to you. The incentives are.

▪Four changes to your retention file before the next resignation lands

1. Read the ranking as a list of labels, not a list of causes. Salary sitting below relocation tells you what departing nurses are comfortable saying. It tells you nothing about what pay is doing to the nurses who stayed. Stop citing a rank position in board materials and cite the underlying behavior instead.
2. Interview the people who stayed, on a schedule. Structured stay conversations at 90 days, at the one-year mark and at eighteen months reach the 43% while they are still yours. That population is larger than your annual departures by a wide margin, and it is the only group you can still act on.
3. Move the exit conversation off your own payroll, or at minimum off the manager’s line. A third-party or fully anonymous post-exit survey thirty days after the last shift surfaces different answers than the conversation held in the building during the notice period. Compare the two for one quarter before deciding which one your retention strategy should believe. The playbooks for running both sit in the Nurse Retention Center.
4. Track intent as a standing metric, not a survey. Add one question to your engagement pulse — how likely are you to be working here in twelve months — and trend it by unit and by tenure band. Turnover tells you what already happened, and by the time it moves, the decision was made two quarters ago.

▪The nurses still on your schedule are the only ones you can still keep

The hospitals getting this right have not thrown out the exit interview. They demoted it. It is a closing formality now, and the real signal comes from the nurses still on the schedule.

That is the harder program to run, and it is the only one that reaches anyone in time.

The Workforce Intelligence Center tracks what your market is paying and how fast your competitors are filling, which is the floor your own nurses are measuring against while they decide whether to keep showing up. Reply to this edition and I will show you what that floor looks like where you hire.

Next Monday: what a quarter of your nurses are absorbing on shift, why so little of it reaches your incident data, and how it comes back to you months later wearing a different label in your turnover file.