Onboarding a Logistics Hire So They Are Still There at 18 Months

There is a specific kind of hiring failure that does not show up in any report. The search worked. The candidate was strong. They accepted, started, and nine months later they resigned — or worse, they stayed and never became the operator you were paying for.

That outcome is expensive twice over. You pay the vacancy cost again, you pay the fee again, and the desk runs degraded for the second time in a year. And unlike a bad hire, this one is almost entirely self-inflicted, because the decisions that produce it are made in the first two weeks, by things that cost nothing.

What the churn numbers actually look like

The U.S. labour market is not especially volatile right now. The Bureau of Labor Statistics put total separations at 5.1 million in August 2026 against 5.2 million hires, with a quits rate of 1.9% and 7.1 million open positions. By the standards of the last five years, that is a calm market.

Logistics is not calm. The Peerless Research Group survey behind this year's supply chain and logistics hiring outlook put average turnover at 11.6%, with about one in five companies reporting no departures at all. Read those two numbers together and the picture sharpens: the churn is not evenly spread. A minority of employers are losing nobody, and the rest are losing more than one in ten people a year.

The same survey names the three reasons people go: compensation pressure, limited advancement, and the pull of a new challenge somewhere else. Only the first is about money, and even that is usually about relative money rather than absolute.

What that means for you is simple. Turnover in this industry is not weather. It is a result, and the employers producing a different result are doing identifiable things.

Day one is about access, and nothing else

The single most common onboarding failure in freight is a new operator sitting for three days without a system login.

It sounds trivial. It is not. A new hire has exactly one data point about how your company runs in their first week, and it is how your company handled the one task it knew about for a month in advance. If CargoWise, Descartes, Magaya or whatever you run is not provisioned on day one, you have told them something true about the operation, and they will believe it for the rest of their tenure.

What to do: request access the day the offer is signed, not the day they start. Email, TMS, phone extension, carrier portals, customs filing credentials, the shared drive, the reporting tool. Have someone log in as them before they arrive and confirm each one works. Put a printed list on the desk with what they have and who to ask for what is missing.

Week one is about ownership

"Sit with whoever is free" is not onboarding. It is an admission that nobody is accountable for whether this hire succeeds.

Name one person. Not the manager — the manager is in meetings. A peer operator who runs a comparable desk, who is told explicitly that this is part of their job for four weeks, and who gets something in return for doing it well. Their job is not to train on process; it is to be the person the new hire can ask the stupid question without cost.

Most new operators in freight do not fail on process. They fail because they do not know who to ask when a file goes sideways at 4pm on a Friday, so they guess, and the guess becomes a claim.

The scope conversation nobody has

Here is the failure that produces the month-nine resignation.

During the interview, the role was described at its best: the accounts, the lanes, the growth, the autonomy. On arrival, the new hire finds the actual desk — which may be perfectly reasonable, but is different. Volume is heavier or lighter. The hybrid arrangement turned out to mean something else. The "ownership" means executing someone else's decisions. Nobody lied. The description was just aspirational, and nobody reconciled it afterwards.

So the hire does the thing people do: they say nothing, adjust their expectations downward, and quietly start taking recruiter calls around month six.

What to do: in week two, sit down and go through what the job actually is, line by line, against what was described in the interview. Name the differences out loud. "We told you this desk had eight accounts. It has eleven right now because Maria left. Here is when that changes." A new hire can handle a harder job than advertised. What they cannot handle is discovering it alone and concluding that you were not straight with them.

30, 60, 90 — with actual content

Most 30/60/90 plans in freight are a formality. Make them specific and they become the thing that converts a hire into an operator.

• By 30 days: they are running live files on a defined subset — one lane, one mode, two or three accounts — with their files reviewed before they go out. Not shadowing. Running, with a net.
• By 60 days: the net comes off on that subset, and the scope widens. They have met the customers on their accounts, by voice, not just by email. They know who the carriers are and who answers the phone at each one.
• By 90 days: they own the full desk as described in the job. You have had one honest conversation about what is still hard, and you have written down what the next twelve months look like — not vaguely, but as a named next thing.

That last item is the one most employers skip, and it is the one that decides month nine. A capable operator who can see what comes after this desk stays. One who cannot starts looking, and the looking is cheap and anonymous.

The marker at 18 months

Eighteen months is the honest test, not twelve. At twelve months a hire is still inside the window where leaving looks bad on a résumé. At eighteen they are fully productive, fully marketable, and — in a market where your customers are building their own logistics teams and your competitors are short-staffed — fully visible.

So the retention work does not end at 90 days. Three things keep operators past eighteen months, and none of them is a raise:

• Scope that moved at least once. A new lane, a bigger account, a process they own, a junior they mentor. Something changed.
• A pay review you initiated. Not one they asked for. If the first conversation about their market value is one they open, you are already negotiating from behind.
• A manager who is not the reason they are looking. This is the one nobody audits, and it accounts for more departures than compensation does.

What this is worth

Run the arithmetic on your own operation. Take one operations seat, the fully loaded cost of it sitting empty for the time it actually takes you to fill — not the time you think it takes — plus the recruiting cost, plus the ramp period during which the replacement is not yet producing. For most mid-market freight employers that number lands somewhere between a third and a full year of the salary.

Now compare it to the cost of the interventions above. A login provisioned on time. A named buddy. One honest conversation in week two. A written answer to "what comes after this desk."

Early attrition in freight is rarely a talent problem. It is an operations problem wearing a talent problem's clothes, and it responds to the same thing every other operations problem responds to: somebody owning it, with a checklist, on a date.

Sources

• U.S. Bureau of Labor Statistics, Job Openings and Labor Turnover Survey, August 2026 (released 29 September 2026) — 7.1 million job openings, 5.2 million hires, 3.1 million quits, quits rate 1.9%, total separations 5.1 million.
• Supply Chain 24/7, 2026 Supply Chain and Logistics Hiring Outlook, citing a Peerless Research Group survey — 11.6% average turnover; 36% of companies added personnel, 32% instituted hiring freezes, 21% reported layoffs; roughly one in five saw no departures; departures attributed to compensation pressure, limited advancement and the attraction of new challenges.