You can usually tell inside ninety seconds whether you like a candidate. That feeling is worth almost nothing, and it is the thing most hiring decisions for this job are quietly built on.
Think about what an interview actually measures. Two people sit down, both of them wanting the conversation to go well — the candidate wants the seat, you want the seat filled, and everyone is being pleasant on purpose. You hire whoever was best at that. Then you send them out to spend six hours a day talking to people who did not want the conversation, are not being pleasant on purpose, and would like it to end.
Those two skills overlap far less than the interview implies. That gap is why a strong week of interviews so often produces a roster that has emptied out by week three.
Nothing below will let you predict who lasts. No process does. What it does is move the guessing earlier, where it costs you an hour, instead of leaving it in week two, where it costs you a training week and a territory that went unworked.
Write down what the seat needs before you meet anybody
Three lines is enough: the hours this seat really requires, the area it covers, and the one thing a person has to be able to do without help. Do it before the first call, not after the third.
Without it you end up judging each candidate against the last one you saw, which is how a weak Tuesday turns an ordinary Thursday into an offer. A written standard is boring and it is the only thing that stops the drift.
The screen’s job is subtraction
A first call should be short, and its purpose is to lose people. If your job post is doing its job, some of that has already happened. The call finishes it.
So say the hard parts out loud in the first few minutes, before you say anything appealing. You are outside all day in whatever the weather is doing. Most doors will say no, and the good days are built out of a lot of those. Here is how pay actually works, including the range your steady people land in — not the best week anyone on the team has ever had.
Every one of those facts is going to be discovered. The only decision you get to make is whether a candidate discovers them on a ten-minute phone call or an employee discovers them on day four. The first costs you a candidate. The second costs you a hire, the week you spent on them, and the doors nobody knocked while you were doing it.
Someone who is still interested after the honest version is a materially different person from the one who was interested before it, even when it is the same person.
Ask what happened, not what they would do
“How do you handle rejection?” tests imagination, and everybody passes. The answer arrives pre-assembled because every candidate has heard the question before.
Ask for instances instead:
- When was the last time a stranger told you no?
- When was the last time you had to do the same thing over and over with nobody checking on you?
- Why did you leave your last job? And the one before that?
Then use the follow-up that does the actual work: what did you do next, that same day? Invented answers have no next day in them — they stop at the point the story was built to make. Real ones keep going, and the detail comes out without being asked for.
Nervousness is not a red flag. It is the most common false one, and passing on a nervous candidate who has done hard work for years in favor of a smooth one who has not is close to the standard hiring mistake in this industry.
The gap between two conversations tells you more than either conversation
This is the part most processes skip, and it is the most useful thing in this article.
Never decide off a single contact. The space between two conversations is the only stretch of your hiring process that resembles the job — a person alone, with something small to do, and nobody watching them do it.
So end the first call by asking for one small, specific, genuinely easy thing before you speak again. Text me the two-hour window you could start on. Look at the company page and send me one question about it. Confirm Thursday by six tonight.
Then watch three things: whether it happened at all, how long it took, and whether they did the thing you asked for or an approximate version of it. That is most of the job. Getting up and doing a repetitive task correctly with no supervision is the whole daily requirement, and this is the cheapest possible test of it. It costs you one text message.
A candidate who was excellent on the phone and then silent for three days has already given you the answer. Believe that one over the phone call.
Finish at doors, not in a room
The last step should be an hour on a street with a working rep. Not a trial shift, not free labor — they watch, and they walk. Tell them plainly that it is a two-way look, because it is, and because a person who backs out after that hour just saved you both a month.
They will not be good at it. That is not what you are watching. Watch what is different at door forty from door four — still asking questions, or gone quiet and walking a step behind. Watch what they do in the thirty seconds after a bad door, which is the part of this job nobody coaches. Watch whether they ask the rep anything, or only ask you.
Run it the way you would run a coaching ride-along, with the same rule: it is not your hour to perform in. And ask afterward what they noticed. The candidate who noticed something specific has been paying attention for an hour, in a situation with nothing in it for them yet.
Two flags worth having
Most of the things managers screen on — no sales background, young, quiet, a patchy resume — are either trainable or beside the point. Two are worth keeping.
Every departure was somebody else’s fault. One bad manager is an ordinary fact of a working life and says nothing. Three in a row, told the same way, is a pattern, and you are going to be in it by the fall.
They never ask how pay works. Someone about to work a commission seat who does not ask how commission works has not thought about being there long enough for it to matter. The candidates who press on this — when it pays, what happens to a canceled sale, whether there are deductions — are the ones taking the job seriously. If your plan has chargebacks in it, this is also your opening to have that conversation before the first sale rather than after the first deduction.
Say no faster than is comfortable
Short-handed managers hire people they have privately already decided against, then spend a training week proving themselves right.
It happens because the two costs are not visible on the same schedule. An empty seat is in front of you every single morning. A wrong hire is paid for later and in pieces — a week of your attention, a territory worked badly, the reps who watched you keep someone who was not trying. The visible cost wins that argument every time unless you make yourself write the other one down before you decide.
You will still hire people who leave, and you will still turn down somebody who would have been good. None of this is a filter that catches everything. What changes is where the finding out happens, and how much it costs when it does.
90-Day D2D Recruiting & Retention Playbook
This article ends at the phone call, and the phone call is the page in the 90-Day D2D Recruiting & Retention Playbook worth printing first: the fifteen-minute phone screen, one sheet per candidate, with the questions to read off it and the red flags to tick while they talk.
If you are further back than this and still short on people to screen, start with recruiting with no ad budget instead — a good process applied to two candidates is not a process, it is a coin toss with extra steps.