How Long Should You Stay at a Job?

How long should you stay at a job? There is no single number that works for everyone, but the number people quietly compare themselves against just moved. The median US worker had been with their current employer for 3.9 years in January 2024, down from 4.1 years two years earlier and the lowest figure the Bureau of Labor Statistics has recorded since January 2002. If your own resume looks short next to a parent's or a mentor's, the baseline itself has been shrinking for two decades, not just your own choices.

How long should you stay at a job, according to the data

That national median already answers the literal question with an honest "it depends," but the trend line is worth sitting with. Tenure has been drifting down for years, and the drop from 4.1 to 3.9 years between the 2022 and 2024 surveys is the sharpest two year move in that direction since the series began falling. A shorter run at a job is not a personal outlier anymore. It is close to the national norm.

Why the number splits so sharply by age

The single median hides a wide gap. Workers aged 25 to 34 had a median tenure of 2.7 years, against 9.6 years for workers aged 55 to 64, more than three times as long. Some of that gap is simple career stage rather than restlessness. People early in a career are still finding the role and employer that actually fit, and a run of two to three year stints is how most careers look while that search is happening. On its own, it is not the warning sign it used to read as.

Why sector moves the number almost as much as age does

Where someone works shifts the median nearly as much as how old they are. Public sector employees had a median tenure of 6.2 years, against 3.5 years in the private sector, and workers in leisure and hospitality had the shortest tenure of any major industry the survey tracks. Comparing your own private sector tenure against a relative who spent a career in government, or against the flat national average, is not a fair comparison to make against yourself.

What employers are actually screening for instead of years

None of this means tenure stopped mattering, but what a hiring manager is trying to learn from it has narrowed. A short stint used to read as a risk signal mostly because there was little else to check quickly. A work sample, a specific checkable claim, or a completed assessment predicts hiring outcomes better than years of tenure or a listed skill, and a process built around checking those directly needs tenure less as a stand in for judgment. The lower the national median goes, the less any single employer can afford to treat two years in a seat as proof of anything by itself.

What is actually worth weighing before you leave

Three questions matter more than a year count. Are you still learning something in the role at a pace you would notice. Is the specific instance of judgment you would put on a CV still growing, or has it gone flat. And does leaving trade a real, known problem for an unknown one, rather than for a genuinely better guess. None of these turn on a number of years. A role that stopped teaching you anything at 14 months is a different case than one still stretching you at 30 months, even though the second number is larger.

The honest limit

The tenure data describes a national trend, not your specific situation. A 3.9 year median does not mean two years is automatically safe and five years is automatically stalled, and it says nothing about what your industry, your household or your own trajectory actually need right now. What it does show is that shrinking tenure is common enough that a shorter work history is not, by itself, something to explain away in an interview. What a new employer checks is what you did with the time you had, not how much time it was. The six domains the check is built to surface do not accumulate by the calendar. They accumulate by what you can point to.