The first ninety days: onboarding an operator so they are still there in year two

The hiring conversation gets all the attention and the retention conversation decides the outcome. A senior hire who leaves at month fourteen has cost you the fee, the vacancy, the second search, and roughly two years of whatever the role was meant to deliver.

With this population the failure is unusually quiet. They do not complain, they do not disengage visibly, and the first clear signal is frequently the resignation. So the useful work is front-loaded.

Week one: say the three things nobody says

“Here is what you can decide without asking.” Write it down. A list, with a dollar figure on it. This single artifact does more for retention than anything else on this page, because the most common reason these hires leave is discovering in month three that the authority they thought they had does not exist.

If the honest list is short, a short list is fine. What is not fine is leaving it unstated and letting them find the edges by walking into them. Somebody who is told “you can commit up to fifty thousand, and anything touching the top three customers comes to me first” will work happily inside that. Somebody who has to discover it by being overruled in front of peers will start looking.

“Here is what you are not expected to know yet.” Name it specifically — the commercial terms with major customers, why the ERP is configured the way it is, the history between two departments. A career where figuring it out is the job produces people who will not raise a hand, and an open-door policy is not an invitation to somebody operating on that norm. Naming the gaps gives explicit permission that a general offer of help does not.

“For your first sixty days, find the problems and do not solve them yet.” Counterintuitive, and it is the single most useful instruction you can give.

An operator arriving in a mediocre process will fix it — correctly, quickly, and frequently before the organization has agreed it was broken. The fix is right and the sequencing is wrong, and what gets remembered is that the new person went around three people in their second month. That memory outlasts the improvement.

State the intent clearly and they will execute it, because clearly stated intent is precisely the thing this population is best at following. Then at week nine you get both the list of problems and the organizational permission to act on it.

Day thirty: ask for the read-out

Schedule it in week one so it is an expectation rather than an intervention.

Ask for three things in writing: what is working that we should not touch, what is broken, and what did you expect to find that is not here. The third question is the valuable one and almost nobody asks it — it surfaces the gap between how you described the company and what it actually is, which is information you can get from nobody else and only in this window.

Two rules for receiving it. Do not defend. The read-out is worthless if the first response is an explanation of why each thing is that way; they will calibrate and the next one will be bland. And respond to every item, even if the response is “we know, and it is not this year’s priority.” An item that vanishes teaches them the exercise was decorative.

Assign a peer, and make it compulsory

Not a mentor — a peer. Somebody at the same level with no reporting relationship, required to have a weekly conversation for the first quarter.

The requirement matters. An optional standing invitation will not be used by somebody whose professional norm says asking for help is a failure of initiative. A recurring meeting they did not request removes the need to ask.

What this buys you is the political map. Which relationships are load-bearing, who has history with whom, which meeting is the one where decisions actually happen. All of that is learnable in a month with a guide and takes a year to reconstruct alone, and the year version involves walking into several of them.

The four signals somebody has already decided

By the time these appear the decision is usually made, but not always, and a direct conversation in the week you notice one is sometimes enough.

They stop raising problems. The most reliable signal and the easiest to misread as settling in. Somebody who brought you three things a week in month two and nothing in month five has not run out of things. They have concluded that raising them does not lead anywhere.

They start executing exactly to spec. Doing precisely what was asked and nothing beyond it. In this population that is a significant behavioral change, and it usually means they have stopped treating the organization’s problems as theirs.

They stop asking about the future. No more questions about next year’s plan or how the function should be structured. People invest in a future they expect to be present for.

The travel appetite changes. Somebody who was happy to be on a site every other week suddenly is not. Sometimes this is genuinely about family. Sometimes it is disengagement, and the two are hard to tell apart without asking.

The retention lever that is not money

If somebody is leaving at month fourteen, a counteroffer will usually not hold them, and if it does it holds them for about nine months.

The thing that actually retains this population is scope that grows. Not a title, not a raise — a bigger outcome to own. A career where advancement meant more responsibility does not translate cleanly into an environment where advancement means a different title and a similar job, and somebody who cannot see the next increment of scope will find one somewhere else.

The practical version: in the first quarter, have a conversation about what they would own in year two if this goes well. It does not need to be a commitment. It needs to be a direction, and the absence of one is read as a ceiling.

What we do, and what we cannot

We stay engaged with both sides after the start date, and this is the part of an engagement clients use least and benefit from most. Practically it means we are talking to your hire during a period when they will tell a third party things they will not yet tell you, and we will bring the substance back to you.

What we cannot do is fix a role that has no real authority, or a compensation decision made on the wrong reference, or a founder who is not actually ready to hand over the list. Those are decisions the company makes before anybody starts, and no amount of post-placement support recovers them.

Which is why the onboarding call asks uncomfortable questions about authority before the search begins rather than after the hire.