Why Your Best Salesperson Is Probably the One You Just Hired

The conventional wisdom in indie retail is that your best salesperson is the one who's been there the longest. They know the inventory, they know the regulars, they know the back stories. They are also, in my experience running a liquor store and now talking to dozens of operators every month, often the worst at converting a customer who walks in without a clear plan. The new hire - the one who started six weeks ago and still has to look up the rye whiskeys - is frequently the better closer. And once you see why, it changes how you think about training entirely.

The veteran's problem

A salesperson with two years on the floor has accumulated three things that quietly hurt their conversion rate.

The first is a script. Two years of answering the same five questions teaches you to answer them on autopilot. Autopilot is fast, and fast is good - except when the customer's question is actually a different question that just sounds like the standard one. The veteran answers what they think was asked. The new hire actually listens, because they don't have a script yet.

The second is cynicism. After two years, you've seen every type of customer. The browsers who never buy, the tire-kickers who ask about every bottle and leave with a six-pack of the cheapest beer, the regulars who only want what they came in for. The veteran has internalized which conversations are worth investing in. They route effort toward customers who look like buyers, and they're often wrong about which customers those are.

The third is store loyalty in the wrong direction. The veteran has favorites - bottles they like, shelves they curated, items they personally picked for the store. They will steer customers toward those items even when something else would suit the customer better. The new hire has no favorites. They steer toward what the customer actually said they wanted.

None of these are failures of effort or attitude. They're side effects of experience. They show up in every category I've watched - liquor, wine, convenience, specialty foods - and they get worse the longer the salesperson has been there.

What the new hire has

The new hire is doing something specific that's invisible to most managers. They're asking clarifying questions. They have to. They don't have the inventory memorized, they don't recognize the regulars, and they're afraid of giving wrong information. So they ask. "What kind of red do you usually drink?" "Are you cooking with this or drinking it?" "Is this for tonight or for the cellar?"

Those questions are the entire job of a good salesperson, and the veteran has stopped asking them.

The new hire also moves slower. Slow is bad for throughput - which is why managers eventually train it out of them - but slow is good for closing rate. A customer who feels rushed buys what they came in for. A customer who feels listened to buys what they actually want, which is usually more expensive than what they came in for.

The third thing the new hire has is no defensive interpretation of customer behavior. When a customer says "I don't know much about whiskey," the veteran hears "this person isn't going to spend money." The new hire hears "this person needs help" - and helps them, and ends up selling a $80 bottle to someone the veteran would have routed to the well.

Why this matters more in 2027

Customer expectations have shifted. The category most people compare retail to now isn't the store next door - it's their phone. Whatever they did online before walking in set a baseline for the level of personalization they expect. When they ask a vague question in person, they expect the kind of follow-up they'd get from a website that knew their browsing history.

The new hire instinctively delivers that. The veteran often doesn't. The gap was always there but it's wider now because the customer-side baseline keeps rising.

This is also where in-store AI starts to fit, but not in the way most vendors pitch it.

AI as continuous training, not staff replacement

The pitch I hear from competitors - and the pitch I refuse to make - is that an AI kiosk replaces the conversion that a well-trained salesperson would have done. That's wrong on two levels. The kiosk doesn't replace the salesperson. The customer wants the human. What the kiosk replaces is the moment when the customer needed help, the salesperson was busy or unavailable, and the customer left without buying anything.

But there's a second use of the same tool that gets undersold. The AI kiosk is, almost incidentally, a transcript machine. Every interaction it has - what the customer asked, how it answered, whether they bought - is a training record. Reviewed weekly with your staff, it's a continuous training loop that no LMS or annual seminar can match.

You sit down with your veteran salesperson. You show them the transcript of a kiosk conversation where a customer described a flavor profile, the kiosk recommended three options, and the customer bought the most expensive one. You ask: "How would you have handled that?" The conversation that follows is the training. You're not lecturing. You're showing the veteran what their best self looked like two years ago, and what they've stopped doing.

The same trick works the other way. The new hire watches transcripts and builds inventory knowledge faster than any printed cheat sheet. They learn product positioning by reading what worked. They learn what questions to ask by reading the ones the kiosk asked.

This is what we mean when we describe Remi as a baseline service layer. It's not a replacement. It's the floor. Your staff is the ceiling. The training loop between them is where the actual revenue lift lives.

What to do about it operationally

Three things, in order of how cheap they are to do.

One: have your new hires shadow nobody for the first week. Most stores have new hires shadow the most experienced salesperson, who teaches them the script. Don't do that. Let the new hire serve customers themselves, with a manager available for questions. They'll be slower for two weeks. They'll convert better immediately, and you'll have a record of what they're actually saying that you can build training around.

Two: review one customer interaction per week with your whole staff. Doesn't matter if it's a kiosk transcript, a security camera clip with audio, or a story your cashier told you. Make it a ritual. Discuss what worked, what didn't. Don't grade. Just notice. Veterans will start to hear themselves.

Three: pay closing rate, not transactions per hour. This is the structural one and it's the hardest. Most retail comp structures reward speed, which trains your staff into the autopilot script. If you can comp on closing rate or basket size, you're rewarding the new-hire behavior, and your veterans will gradually rediscover it.

The deeper point

The retail industry spends a lot of money on training and not very much on un-training. The shift from a six-week new hire to a two-year veteran is mostly an un-training problem - they're learning to do less. AI tools, used correctly, are an un-training intervention. They make the script visible so you can break out of it.

If you want to see what this looks like in practice, the demo page walks through a few real customer interactions and what they tell you about staff behavior. The About page has the founding story, which is basically a long version of this argument.

Frequently asked

Isn't this an excuse to keep new hires underpaid?

No. The argument is that new hires are doing something specific - listening, asking, slowing down - that's worth identifying and protecting. The conclusion isn't that they should stay junior forever. It's that veterans should keep doing what new hires do.

What if my veteran really is my best salesperson?

Some are. The good ones never lost the listening behavior. They're rare, and they're usually self-aware about it. If you have one, study them and let your other staff shadow them. The point of this article isn't that all veterans are bad - it's that experience is not automatically an asset, and most stores treat it as if it is.

Does this apply to part-time staff?

Even more strongly. Part-timers don't accumulate the autopilot script as quickly because their reps are spread out. They tend to keep the listening behavior longer. Many of the operators I talk to have noticed that their best Saturday closer is someone who only works one day a week.

How does the AI fit if I don't have a kiosk?

Same principle, different tool. Anything that records and lets you review customer interactions - even just notes from each shift - serves the same continuous-training function. The kiosk is convenient because it generates the records automatically. It's not the only path.

Is there a category where this is wrong?

High-trust, expert-driven categories - high-end wine, premium cigars, niche specialty foods - lean more on veteran knowledge. But even there, the veterans I've seen do best are the ones who kept asking clarifying questions. The pattern holds; the magnitude shifts.