The Job Posting Is the Smallest Line When Someone Leaves in Month Five

When an employee who started in January resigns in June, the cost that gets discussed in the room is the job posting, and occasionally the recruiter's fee if one was used. That is the smallest line on the list by a wide margin, and treating it as the headline distracts from a total that in most skilled roles runs to several months of the position's salary. Counted honestly, the expense falls into four buckets, only one of which ever arrives as an invoice, which is precisely why the other three go unexamined year after year while the same pattern repeats.

What You Already Spent and Will Not Get Back

Everything invested in making this person productive is written off on the day they hand in the notice. That means recruiting time first: writing the ad, screening, interviewing and checking references, counted as hours of everybody involved and priced at what those hours actually cost rather than at zero. Then onboarding, meaning paperwork, equipment, accounts and training materials. Then supervised time, which is the line people forget, because a new person's first weeks consume somebody else's hours alongside their own and that somebody is usually the most experienced person in the building.

Add any signing incentive, relocation payment or training course already paid for, and the bucket is closed. The uncomfortable feature of it is that almost all of it is fixed rather than variable, so replacing the person means paying every one of those lines again at full price, with no discount for having done it recently.

The Productivity Gap Is the Number That Never Appears

This is the largest figure in the accounting and the one that never shows up in any ledger. A new hire in most skilled roles reaches full productivity somewhere between three and nine months in, depending on how complex the work is and how well the training was organized, which means somebody leaving in month five was still climbing the curve. The employer paid a full wage for a partial contribution and then lost the stretch where the contribution turns positive and starts repaying the investment.

Then the seat sits empty, and during the vacancy the work is either not done, done by somebody else at the expense of their own work, or done on overtime. All three carry a price, and the middle one is the quietly damaging option, because it converts one vacancy into a slow tax on the people who stayed. After that the replacement starts and the ramp begins again from zero, which is the part that makes an early departure so much more expensive than a departure at month thirty.

The Costs Nobody Assigns to the Departure

Four effects show up in other places on the books and never get traced back. Customer relationships suffer where the role has external contact, because five months was long enough for somebody to become the person a customer asks for by name, and the reintroduction spends goodwill that took a year to build. Knowledge that was never written down walks out too, meaning five months of small operational fixes, workarounds and account details that lived in one head. Errors rise during the coverage period, since people doing two jobs at once make more of them and the mistakes surface weeks later. And departures cluster, because a resignation prompts other people to look around, and the ones with the most options look first.

Why the Second Search Costs More Than the First

Rehiring for the same role within months is harder rather than easier, which runs against the intuition that the work has already been done once. The pool has been used, and the strong candidates from the first round have taken other jobs by now. Anybody who notices the same role advertised twice in six months will ask why, and the honest answer is a difficult sell in an interview. Worst of all, the pressure to fill quickly is higher because the seat is already empty rather than about to be, and hiring under time pressure is precisely the process that manufactures the next early departure.

Where It Goes Wrong, Usually Before Day One

Early departures cluster around a small set of causes and most of them were decided before the person ever walked in. The job was not the job described, whether in hours, duties or the size of a task the ad mentioned in passing. The pay sat below the market and they found out within weeks, from a colleague or a job board. Nobody trained them, so a capable person left to work it out alone concluded they were failing and left before anyone told them otherwise. There was no visible path, which need not mean promotion and often means nothing more than a clear answer about what next year looks like. Or there is a manager problem that predates them entirely, which is the diagnosis worth taking seriously when a role has turned over repeatedly, since at that point the role is not the variable.

Separations and turnover across industries are tracked by the Bureau of Labor Statistics, and holding your own pattern up against the general shape for your sector is a useful corrective before concluding that you have a hiring problem. Many employers who believe they cannot find good people turn out to be losing them at an ordinary rate for the industry, which is a retention question with entirely different answers.

The Cheap Interventions That Move the Number

What works is small and early, which is convenient given that what works is also inexpensive. Describe the job accurately in the ad, including the parts people dislike, because candidates who accept anyway do not later resign over them. Check pay against the market annually rather than at the moment somebody resigns, since a counteroffer is the most expensive possible way to discover that a range has drifted. Name one person responsible for training and give them the time to actually do it. Hold the thirty and sixty day conversations, because most early resignations are visible weeks before they arrive if anybody is looking. And ask departing employees what happened a month after they leave rather than on the last day, because the two answers are rarely the same and only one of them is useful.