This problem has become sharper in every market in recent years: hardly anyone answers the advert, those who do often don't turn up on the first day, and those who do start move on two months later. At that point the shopkeeper typically focuses on the search — yet the bigger stake is on the other side.
It is worth working out once what a leaver costs. The price of the advert isn't the point; the point is that your time goes into the weeks of training, that there are more errors and slower service at the till, and that while there is nobody, you step in — often on your day off. Against that, the few things that would make someone stay are usually very cheap.
This article falls into two parts: where and how it is worth looking if you genuinely need someone, and what can be done so that you don't have to look again in six months. The second part is the more important one.
What a leaver actually costs
It is worth writing this down once, because in most shops it is an invisible cost. It doesn't appear on any line anywhere, yet it is real — and far larger than what ought to be spent on keeping people.
| Item | What happens? | Whose time? | How visible is it? |
|---|---|---|---|
| Searching | Advert, phone calls, conversations | Yours, typically after closing | Barely — there is no invoice for it |
| Training | For the first days two of you do one job | Yours or an experienced colleague's | Invisible, but the dearest item |
| Mistakes | Mis-keyed sales, slower service, discrepancies | The customer's too — queue, irritation | Only in the month-end figures |
| The gap period | While there is nobody, you step in | Your day off | Not in money, but in exhaustion |
| Effect on the team | Those who stay work more, the mood drops | Everybody's | In the next resignation |
Where to look — and where not to
For a small shop the big job boards generally perform poorly: the advert is lost among the chains' offers, and it typically reaches people who don't live nearby. And distance is the main reason for leaving: someone travelling forty minutes will move closer at the first opportunity.
| Channel | How well does it work? | Why? | What to watch |
|---|---|---|---|
| A sign in the window and on the counter | Very well | People living nearby see it, and they know the shop | Put on it when and how to get in touch |
| Asking your regulars | The best hit rate | You know them, and they know you | Ask whether they know anyone |
| A referral from an existing member of staff | Outstanding | Nobody refers someone they'd be embarrassed by | Worth thanking them if it works out |
| A local community group | Well | It reaches exactly the people nearby | State the hours too, not just the job |
| A large job board | Poorly | Lost among the bigger offers | Only once the others have brought nobody |
The wording of the advert matters too, and there is a common mistake here: everyone writes "shop assistant wanted". The candidate learns nothing from that. Write what actually interests them: which times of day they would work, how many hours, roughly what pay to expect, and when they can call. An advert that says this brings fewer but far more usable applicants.
What to look for in the conversation
Two things really matter in small-shop work: is someone reliable, and can they deal with people. Handling the till, the products and the tidying can be taught in a few days. Punctuality and basic kindness cannot.
- Did they arrive on time for the conversation? It is a single data point, but the best predictor of future attendance there is.
- How far away do they live? A long journey is the commonest reason for leaving — this can be known in advance and cannot be fixed afterwards.
- Why did they leave their previous job? What matters isn't the content of the answer but whether it is matter-of-fact or everyone else was to blame.
- Do they know which times of day they would work? Go through it concretely, because a surprise later becomes a resignation after two weeks.
- How do they behave when a customer comes in during the conversation? That shows in half a second what half an hour of talking doesn't.
- Do they ask anything at all? Someone who asks no questions typically isn't planning to stay long either.
- Is there something they clearly won't take on? Better to know now than at the first Sunday shift.
One practical method helps a great deal: invite them into the shop for half an hour during a busy period, and stand beside the counter. Not as a trial shift, but as a conversation. In half an hour you can see whether the bustle bothers them, whether they look at whoever walks in, whether they ask. That says more than any CV.
The first two weeks decide it
The bulk of early departures don't happen in the third month but in the first two — and nearly always for the same reason: the person was left alone. They were thrown in at the deep end, everyone was busy, asking felt awkward, and after two weeks they felt they would never learn it.
- 11. Have someone beside them all of the first dayDon't let them stand behind the counter alone, even on a quiet day. Let the first solo sale be one where there is somebody to turn to — that day decides the months to come.
- 22. Tell them in advance what they will know by the end"Day one is the till and greeting people. Day two the chiller and pricing. On day three you open on your own." Someone who knows where they are heading doesn't feel incompetent.
- 33. Write it down, don't say itIn the first week nobody can remember twenty rules. One sheet of paper under the counter — opening, closing, who to call, where things are — is worth more than saying it three times.
- 44. In the first week, ask every day what was hardTwo minutes at the end of the shift. That is how you find out what they didn't dare ask. Those two minutes are why they won't hand in their notice without a word later.
- 55. Give the rota in advanceFrom the first week there should be a rota visible two weeks ahead. Uncertainty drives people away just as fast as low pay.
- 66. Sit down for ten minutes at the end of the second weekWhat works, what doesn't, is anything bothering them. If something is wrong, at this point it can still be said. A month later it comes out as a resignation.
„Most people don't leave over pay. They leave because they don't know when they are working, and because nobody stood beside them in the first week.”
Why they leave — and what keeps them
Pay matters, but it is rarely the first reason. Most departures come down to three things: the rota is unpredictable, the person is left alone, or they feel treated unfairly. All three can be dealt with more cheaply than a pay rise — and they last longer than one, too.
| Reason | How it shows | What it costs to fix | What to do |
|---|---|---|---|
| An unpredictable rota | "I can't plan anything" | Nothing — only organising | A rota issued two weeks ahead |
| Being left alone | Makes mistakes quietly, doesn't ask | A few hours in the first week | Have someone to turn to in the early days |
| Unfairness | "I always get the weekend" | Nothing — only transparency | A rotating order, on a visible rule |
| No feedback | Doesn't know whether they're doing well | Two minutes per shift | Say it out loud when something was good |
| The pay really is low | They get a concrete offer elsewhere | This is the only one that costs money | Look at the local level honestly |
An honest note on pay: wages won't keep someone who is left alone, but being underpaid can't be made up for with a good atmosphere either. It is worth checking once a year what similar work pays nearby, and if the gap is serious, dealing with it — before it arrives in the form of a concrete offer.
What costs nothing and still keeps people
- A predictable rota two weeks ahead. This is the strongest retention tool there is, and it costs nothing — only sitting down in time to do it.
- Fair weekends and holidays. A rotating order, on a visible rule. If the same person always gets the worse end, sooner or later they leave without saying so.
- Feedback said out loud. "You handled that customer well today." One sentence at the end of the shift — in most places this is missing entirely.
- Clear boundaries about what is and isn't allowed. An unclear rule is worse than a strict one: nobody knows when they are getting it wrong.
- A proper framework for staff consumption. With a rule and easy recording, it never becomes an awkward situation for anyone.
- Tools that don't work against them. If the till is slow and the products have to be known by heart, every shift is a struggle; if the system helps, the work is easier.
- Using their name, and a human connection. In a small team this isn't a soft factor but the most important one.
What not to do
- Don't hire just anybody out of desperation. A bad hire costs more than an empty position — and takes those who stay with them.
- Don't promise a rota you can't keep. It comes out in the second week, and from then on they believe nothing you say.
- Don't leave them alone on the first day. This is the commonest and costliest mistake in the whole process.
- Don't train them on everything at once. Day one is the till and greeting people; the rest can come later.
- Don't hide the pay in the advert if it will come out anyway. You are only wasting your own time with the applicants.
- Don't speak ill of former colleagues to the new one. They will know exactly what you'll say about them when they leave.
- Don't let someone who resigns go without a word. The honest exit conversation is the cheapest lesson you can get.
Summary
- Keeping people is cheaper than searching — the real cost of a leaver is training, mistakes and your time.
- Look nearby: a sign, your regulars, referrals, a local group — the job board comes last.
- Hire for reliability; the till can be taught, punctuality cannot.
- The first two weeks decide it: have someone to turn to, and make the path of the training visible.
- A predictable rota, fair weekends, feedback said out loud — three of the four reasons for leaving aren't money.
- Talk to the person leaving: it is the cheapest lesson for the next hire.
Frequently asked questions
Frequently asked questions
Where do I even find a shop assistant if nobody answers the advert?
Start with the closest circle: a sign in the window and on the counter, asking your regulars, a referral from existing staff, a local community group. These are free and have a far better hit rate, because they reach people who live nearby and already know the shop. Leave the big job board until last: there your advert sits among the chains' offers and typically reaches people further away — and a long journey is the commonest reason for leaving.
What should I put in the advert?
Exactly what the candidate wants to know: which times of day and how many hours they would work, roughly what pay to expect, where the shop is, and when they can call. "Shop assistant wanted" says nothing on its own, and because of that either nobody applies, or people do whose hours turn out not to fit at the first conversation. A concrete advert brings fewer but far more usable applicants.
Should I look for someone experienced or someone I can train?
In a small shop reliability usually matters more than experience. Handling the till, the products and the tidying can be taught in a few days — punctuality, showing up and basic kindness cannot. A punctual person willing to learn is a better choice without experience than someone who knows everything but is a doubtful starter. In practice the best signal is whether they arrived on time for the conversation.
How can I judge in half an hour whether they'll be any good?
Invite them into the shop during a busier period and talk beside the counter. See whether the bustle bothers them, whether they look at a customer coming in, whether they ask anything. Ask how far away they live and why they left their last job — what matters isn't the content of the answer but whether it is matter-of-fact. That half hour shows more than any CV, because you see it in the real situation.
Why do they leave after two months?
In most cases because they were left alone in the first week. They were thrown in at the deep end, everyone was busy, asking felt awkward, and after two weeks they felt they would never learn it. The other two common reasons are an unpredictable rota and weekends that feel unfairly shared out. Pay usually comes third or fourth — although if you are seriously below the local level, it moves up.
What does it actually cost when someone resigns?
Far more than the price of the advert, it just doesn't show on one line. It includes the time spent searching, the weeks of training — when two of you do one job — the till errors and slower service, the period when you step in on your day off, and the load on the colleagues who stay. If you write this down once, it usually turns out that the few organisational steps that keep people cost a fraction of it.
What if I can't pay more?
Then concentrate on the other four reasons, because they aren't about money. Give a predictable rota two weeks ahead, share weekends and holidays fairly on a visible rule, have somebody to turn to in the early days, and say it out loud when someone did something well. Together these are worth more than a few per cent in many places — though if your pay is seriously behind the local level, that eventually can't be substituted with anything else.
How do I train someone when I have no time for it?
Not on everything at once. Day one is the till and greeting people, day two the chiller and pricing, day three opening alone — and say in advance that this is the plan. Write the essentials on one sheet (opening, closing, who to call, where things are), because in the first week nobody can hold twenty rules in their head. The time you spend on the first day is the best-returning investment in the whole process.
Is a trial period needed, and what should I do with it?
The legal rules on probation differ by country, but from an operational point of view two things are generally true. One: at the end of the first two weeks sit down for ten minutes and have both of you say what works and what doesn't — at that point it can still be fixed, whereas a month later it would be a resignation. Two: if lateness or not turning up is regular in the first weeks, it doesn't improve by itself; better to settle that early than carry it for six months.
What should I do when someone resigns?
Sit down with them honestly and ask what it was that made them leave. Say you aren't trying to talk them round but to learn from it — people say far more then. The answer you get here is the cheapest lesson available, because the same reason would very likely take the next person too. And if you part on good terms, they often come back, or refer somebody in their place.
Is it worth employing students or pensioners?
In many small shops this is a workable solution, especially for covering peak hours, but in different ways. Students are typically flexible, though they drop out during exams and turnover is higher — it is worth training two people for the same shift. Retired colleagues often bring outstanding experience and reliability, and frequently want exactly those stable, predictable few hours. The legal and tax conditions of employment differ by country, so that is worth settling locally.
How do I trust someone I have only just hired?
Not all at once and not with everything — build it gradually. In the first week they can serve customers, but the daily close and the turnover figures stay with you; keep the sensitive pages of the system locked with a password until trust develops. This isn't mistrust but protection for the other person too: someone with no access cannot be called into question. As time passes, the circle widens by itself.
A new colleague can stand at the counter on day one.
In Boltom App a photo and a short code are enough to serve — nobody has to know the products by heart. Days of training instead of weeks, and fewer people give up because of it.
- A photo and a short code for every product — it doesn't depend on memory.
- 21 languages: your colleague uses the system in their own language.
- Sensitive pages stay locked with a PIN while trust is being built.
No card needed · 2 minutes
