Hiring your first employee without it going wrong

You are working seventy hours and turning down work. So on a Tuesday, in the middle of a bad week, you decide to hire someone. A friend recommends a cousin. You agree a wage in a two-minute conversation, they start on Thursday, and nothing is written down anywhere.

Three months later you are still working seventy hours, because now you do your job and you supervise badly. You cannot say whether they are doing well, because you never said what well meant. You suspect it is not working, and you do not know how to end it without a mess, so it runs for another year.

That story is not about a bad employee. It is about a job that was never defined, training that never happened, and a standard that was never stated. Almost every failed first hire fails on those three, and all three are yours to fix before anyone starts.

By the end of this guide you will know whether you can actually afford someone, which job to hire for (which is usually not the job you think) how to test a candidate before you commit, what has to be written down, how to train them in thirty days, and how to tell within weeks whether it is working.

(Written for any country. Money appears as plain numbers, read them in your own currency. Employment law is genuinely different everywhere: contracts, minimum wage, social contributions, notice periods, dismissal process. Where you see a legal step, treat this as the checklist of what to ask about locally, not as the law itself.)

Can you actually afford one?

The wage is not the cost. Work out the real number before you go any further.

Monthly wage                                      ________
+ Employer contributions, payroll taxes,
  insurance required where you are                ________
+ Tools, equipment, phone, uniform, space
  (spread over 12 months)                         ________
+ Recruitment and any training cost               ________
= CASH COST PER MONTH                             ________

Then add the cost nobody budgets: your own time. Expect to spend a significant part of your working week for the first four to eight weeks training and correcting. If your hourly rate is 800 and you lose ten hours a week for six weeks, that is 48,000 of your capacity, spent before the person is productive.

Now the test that decides it. An employee has to either free up hours you can sell, or produce sales themselves:

Value created = hours of yours freed × your hourly rate
                (or) sales they generate × your gross margin
Value created must comfortably exceed the cash cost per month.

"Comfortably" matters, because the first months run at a loss while they learn. If the sums only work when the new person is fully productive from week one, they do not work.

Two more checks before you commit:

• Can you pay them in your worst month? Wages are the one cost that cannot be delayed and cannot be negotiated down when trade dips. Look at your slowest month last year and confirm the wage fits inside it.
• Is the problem actually a people problem? Sometimes seventy hours is a systems problem: no prices written down, no reorder levels, everything routed through you, no processes anyone else could follow. Hiring into chaos multiplies the chaos and adds a wage to it.

Hire for the right job, which is rarely the obvious one

Most owners hire someone to do the interesting part of their own job, and stay buried under the boring part. Do this exercise before you write any advertisement.

For two weeks, keep a running log of everything you do, in fifteen-minute lumps. Rough is fine. Then sort every line into three columns:

Add the hours in each column. Then hire for column A first, almost always.

It feels wrong. Column A is the cheapest work, and paying someone to do it looks like the smallest win. But column A is easy to teach, easy to check, low risk if it goes wrong, and it frees the largest number of your hours the fastest. Hiring someone for column B before you have proved you can manage anyone is how first hires end badly.

The output of this exercise is a job that already exists on paper, in hours, made of tasks you can describe. That is the thing you are about to advertise, and it is where most people start guessing instead.

The one-page job description

Write it before you speak to anyone. One page, and it is as much for you as for them.

• Job title and who they report to. Even if it is just you.
• The five things they own. Written as outcomes, not activities. Not "help with stock" but "shelves are full every morning before opening and the count sheet is completed daily."
• Hours and days, including how overtime or busy periods work.
• Pay, and when it is paid. A specific figure and a specific date.
• Probation period, and what happens at the end of it.
• What good looks like at 30, 60 and 90 days. Three short lists. This single section prevents most first-hire disasters.

That last part is the one people skip, so make it concrete:

By day 30: opens the shop alone, completes the daily count sheet,
serves customers without help on the top twenty items.
By day 60: handles deliveries and checks them in, manages the
cleaning rota, answers routine messages using the saved replies.
By day 90: runs the shop alone for a full day, spots and reports
stock about to run out.

Finding people worth interviewing

In order of how well they work, for most small businesses:

Referrals from customers and suppliers. The strongest source. People recommend someone whose reputation touches theirs. Ask specifically: "do you know anyone reliable looking for work in [role]?"

Your own past casual or seasonal helpers. You have already seen them work, which is worth more than any interview.

A well-written advertisement in the places your candidates actually look. Local groups, notice boards, trade schools, community pages, whatever applies where you are.

Five lines is enough, and one of them should do the filtering:

[Role] wanted at [business], [area]. Full time, Mon-Sat.
You will be responsible for [two of the five outcomes].
We need someone reliable, on time, and comfortable with customers.
Pay: [figure] per month, paid on the [date].
To apply, send a message saying why this suits you, and include
the word "shelf" so I know you read this.

That last line is not a trick. It tells you instantly who reads instructions, and it will remove a large share of applicants who would have wasted your week.

About hiring relatives and friends: it can work, and it fails in a specific way. The terms are never written down, and correcting them costs a family relationship. If you do it, write everything down more carefully than you would for a stranger, not less.

Interview to learn something, then test

An interview mostly measures how well someone interviews. Keep it short, ask about behaviour rather than intentions, and then get evidence.

Six questions that earn their place:

1. "Walk me through a normal day in your last job." Detail and sequence tell you whether they did the work they claim.
2. "Tell me about a time something went wrong at work. What did you do?" Listen for whether they acted or waited, and whether anything is ever their fault.
3. "What did your last manager think you needed to improve?" The refusal to answer is itself an answer.
4. "What would stop you being here at 7am every day?" Transport, childcare, distance, another commitment. Better discovered now than in week three.
5. "This job involves [the least attractive part]. How do you feel about that?" Say the worst part out loud. People who leave in month two usually did not know.
6. "What do you want to be doing in two years?" You are checking that the job is not obviously a waiting room.

Then the part that actually predicts performance: a paid trial task. Half a day or a day, paid at the going rate, doing the real work. Watch how they turn up, how they ask questions, how they handle being corrected, and what they do when they finish something.

Almost nobody regrets a paid trial and almost everybody who skips one regrets that.

Before you offer: call two references and ask two questions. "Would you employ them again?" and "what should I know to get the best out of them?" The pause before the answer to the first one tells you most of what you need.

The paperwork, and the hour that saves you a year

This is where local law does the deciding, and where getting it wrong is expensive. Treat the list below as the questions to take to a local accountant, labour office, or small business advisor. One paid hour with someone who knows your country's rules is the cheapest insurance in this guide.

• A written contract or statement of terms, signed by both of you, before the first day. Job title, duties, pay, hours, start date, probation, notice, leave, and where they work.
• Registration and contributions. Most countries require you to register as an employer and to deduct or contribute to tax and social security schemes. Find out which apply, the rates, and the deadlines.
• Minimum wage and working time rules, including maximum hours, rest days and overtime.
• Leave entitlements, annual, sick, parental, public holidays.
• Insurance. Employer's liability or workplace injury cover is compulsory in many places and sensible everywhere.
• Payroll records. Keep a payslip and a record of hours and payments for every period, for as long as your country requires.
• Right to work checks where they apply, done for every hire, identically, before the start date.
• Health and safety obligations for your kind of work.

Two rules that survive any legal system. Never pay wages in cash with no record. It exposes you to penalties, it leaves you unable to prove what was paid, and it removes any protection you have in a dispute. And put the probation period in writing, with a review date in your calendar, because probation you forget to use is probation you did not have.

The first thirty days decide everything

Most people who fail in a job were never taught it. Plan the first month before they arrive.

Day one is not "watch me and pick it up". Show them where everything is, introduce them to the regulars, give them one task they can complete successfully by the end of the day, and tell them exactly what tomorrow looks like.

Teach each task in four steps. You do it while they watch. You do it together. They do it while you watch. They do it alone and you check the result. Rushing straight to step four is why owners conclude that nobody can do it properly.

Have them write the checklist. After they have learned a repeated task, ask them to write down the steps in their own words. You correct it, and it becomes the written procedure. It proves they understood it, and it means the next hire trains in half the time.

Check work early and visibly, then taper. Daily in week one, every other day in week two, weekly by week four. Checking is not distrust; it is how you catch a small misunderstanding before it becomes a habit.

Give the correction immediately, privately, and about the action. "The count sheet was not filled in yesterday, walk me through what happened" beats a general complaint three weeks later about attitude.

Say what is going well, specifically. "You handled that customer well when she was angry" teaches far more than "good job", because it tells them exactly what to repeat.

Fifteen minutes a week, forever

Once a week, same time, sitting down, no phones. Three questions:

1. What went well this week?
2. What got in your way?
3. What are the priorities next week?

Write the answers in a book. It takes fifteen minutes and it does three things: it catches problems while they are small, it gives you a record if things go wrong later, and it makes the person feel managed rather than watched.

At the end of probation, actually hold the review. Use the 30/60/90 lists you wrote. If they have met them, say so and confirm the role in writing. If they have not, this is the moment to act, with far more options and far less cost than you will have in six months.

When it is not working

Diagnose before you decide, because the three causes need different responses.

Unclear expectations. They do not know what good looks like, or they think they are doing fine. Very common, and it is your fault, and it is fixable in a conversation. Restate the standard in writing, with dates.

Capability. They understand and cannot do it yet. Ask whether more training or a narrower job would fix it, and give it a defined window, two weeks, four weeks, with specific things to achieve.

Attitude or reliability. Lateness, dishonesty, rudeness to customers, ignoring instructions. This rarely improves with patience, and tolerating it damages everyone else who works for you.

Whatever the cause, do this the same way:

• Talk to them directly and early. Most owners wait months, then explode over something small.
• Write down what was said, when, and what was agreed. Every time. This is the record you will need.
• Give a clear standard and a clear date. "The shop must be open by 7.30 every day. We will review this on the 30th."
• If it must end, follow your local process exactly. Notice periods, final pay, procedure and any required warnings are set by law where you are, and shortcuts here cost more than the whole employment did. Get advice before you act, not after.

Ending an employment properly is not cruelty. Keeping someone in a job they cannot do, while resenting them, is worse for both of you than a clean, fair, well-handled end.

What goes wrong

Hiring because you are drowning, in a hurry. Panic hiring gets the wrong person into the wrong job, and it takes far longer to undo than to do properly.

Nothing in writing. No contract, no job description, no standards. Every later dispute is then a matter of memory, and you will lose it.

Hiring for column B or C before you can manage anyone. Start with the work that is easy to teach and easy to check.

No probation, or a probation nobody reviews. Put the review date in the calendar on the day they start.

No training plan, then blaming the person. If you cannot describe the job in writing, you cannot fairly conclude they are failing at it.

Paying cash with no records. Illegal in most places, and it leaves you with no defence in any dispute.

Hiring a relative on "we will sort it out". The one relationship where written terms matter most, and where they are most often skipped.

Waiting a year to act on a bad hire. The cost is not just their wage; it is the customers, the mistakes, and the good staff who watch you tolerate it.

Forgetting wages in slow months. Wages continue when sales fall. If your cushion cannot cover three months of them, hire part-time or on defined short-term work first.

Your hiring timetable

Start the task log tomorrow morning, before you advertise anything. Most owners who do it discover that fifteen to twenty hours of their week is column A work, which means the job they were about to advertise was the wrong one, and the right one is cheaper, easier to fill and frees more of their time.