43% of New Hires Quit Because the Job Wasn't What They Expected

If you run an onboarding program, you almost certainly have a dashboard. It probably shows completion rates — modules finished, sessions attended, forms signed. Green across the board, most weeks.

Here's the number that dashboard isn't showing you: 43% of employees who leave within their first 90 days say the day-to-day role wasn't what they expected. Not pay. Not their manager. Not the commute. The job itself didn't match the picture in their head.

That's not a small footnote in a workforce survey. It's the single most commonly cited reason new hires give for leaving early. And it's a number your completion dashboard has no way of catching, because completion and expectation alignment are not the same measurement, even though most onboarding programs treat them as if they were.

Quick Answer

43% of new hires who leave within their first 90 days cite a mismatch between the expected and actual day-to-day role as their primary reason — making it the leading driver of early attrition. Standard onboarding metrics like completion rate and attendance don't measure this at all; they confirm that content was delivered, not that the new hire's mental model of the job matches reality. Closing this gap requires a different kind of measurement — one that checks alignment, not just attendance.

What Completion Rates Actually Tell You — and What They Don't

A completion rate answers one question: Did the new hire go through the material? It says nothing about whether they believed what they learned, whether it matched what they expected, or whether they're quietly recalculating their decision to join.

This matters because completion and expectation alignment can move in completely opposite directions at the same time. A new hire can complete 100% of an onboarding program while their confidence in the decision to join is steadily eroding.

Completion tracking exists because it's easy to measure. Expectation alignment is harder to measure, because it requires actually asking, more than once, whether what someone is experiencing matches what they expected. Most onboarding programs default to measuring the easy thing and assume it's a reasonable proxy for the hard thing. The 43% figure is the clearest evidence that it isn't.

As explored in The Job You Hired For Isn't the Job They Think They Joined, this expectation gap forms before onboarding even begins — during hiring — and widens silently unless something is actively checking for it.

Two Different Questions, One Dashboard Only Answers

Why This Number Predicts Something Completion Rates Can't

The reason the 43% figure matters so much is that it's not describing a vague feeling — it's describing a specific, identifiable category of failure: the role, as lived, diverging from the role, as imagined. That kind of mismatch tends to show up early and tends to compound.

This is also why the number is so actionable, once you know to look for it. Unlike vaguer attrition drivers, an expectation mismatch can be checked for directly. You can ask someone, early and specifically, what they expected their day-to-day to look like, and compare the answer against what their day-to-day actually is. Completion rates were never going to surface it, because they were never designed to.

What Measuring Expectation Alignment Actually Looks like

This isn't about adding a satisfaction survey at the 90-day mark — by then, the decision has often already quietly formed. It's about building a small number of specific comparison points, early, that check belief against reality before the gap has time to compound.

That means:

• Asking a new hire, directly, what they think their role involves day to day.

• Comparing that answer against what the role actually requires.

• Checking that comparison more than once across the first 90 days.

• Giving managers a structured way to flag the gap when they see it.

This is one piece of a broader diagnostic approach — one that treats expectation mismatch as falling into a small number of distinct categories rather than a single vague “fit” problem. As covered in It's Not Always a Skills Gap. Sometimes the Mismatch Is Somewhere Else Entirely., the day-to-day role is only one of at least four places this kind of gap tends to open.

Frequently Asked Questions

The explainer breaks down exactly where this 43% comes from and what it predicts.