Should working interviews be paid?

In most cases, if the candidate produces real work that benefits your business, such as serving customers or completing a task you would otherwise pay staff for, they should be paid. Rules vary by country, so check your local labour regulations before deciding, and always confirm pay arrangements in writing before the session starts.

How long should a practical test be?

Keep it as short as possible while still being useful, typically thirty to ninety minutes for a desk-based exercise or up to half a day for a paid trial shift in a hands-on role. If the task genuinely needs longer than that, it is a sign you should be paying for the time and treating it as real work rather than an assessment.

How do I make sure a practical test is fair to all candidates?

Give every candidate applying at the same stage the exact same task, instructions, time limit, and level of support, and score everyone against a written rubric agreed before the test begins. Consistency is what makes the comparison meaningful, so avoid tailoring the difficulty or help offered based on how you feel about a particular candidate.

Can I use a practical test as my first screening step?

It is better used later in the process, after an initial conversation, rather than as the first filter for every applicant. Asking large numbers of early-stage candidates to complete lengthy exercises tends to discourage strong candidates who simply do not have the spare time, and it rarely saves you as much effort as it seems to.

What should I do if a candidate cannot attend a scheduled practical test?

Offer a reasonable alternative time or format where possible, since rigid scheduling can unfairly exclude candidates with caregiving responsibilities, other jobs, or disabilities. If a candidate needs an accommodation to complete the task fairly, such as extra time or adjusted equipment, treat that as a normal part of running a fair process rather than an exception.

How to Run a Working Interview or Practical Test Fairly

A practical test can reveal far more than a conversation, but only if every candidate gets the same task, the same time, and the same shot at showing what they can do.

A resume tells you what someone claims they can do. An interview tells you how well they talk about it. A working interview or practical test tells you something closer to the truth: what actually happens when they sit down and do the work. Done well, it is one of the most useful tools in hiring. Done badly, it can waste a candidate's time, produce unreliable results, and expose your business to fairness complaints.

This guide walks through how to design and run a practical test that is genuinely fair to every candidate, useful to you, and legally sound wherever you happen to be hiring.

What a working interview or practical test actually is

A working interview is any hiring step where a candidate performs real or realistic work instead of just talking about it. That might be a paid trial shift on the shop floor, a short coding exercise, a mock sales call, a design brief, a writing sample completed on the spot, or half a day shadowing a role with some hands-on tasks mixed in.

The point is to reduce guesswork. Interviews reward people who present well under questioning. A practical test rewards people who are actually good at the job, which is not always the same skill. For roles where the day-to-day work is concrete and observable, a well-run test is often a better predictor of success than another round of conversation.

When a practical test makes sense (and when it does not)

Practical tests work best for roles with a tangible, assessable output: a chef, a developer, a copywriter, an electrician, a sales rep, a customer support agent, a bookkeeper. In each case you can construct a small, realistic slice of the actual job and watch someone do it.

They work less well as an early screening step. Asking every applicant to complete a two-hour task before you have even spoken to them is a fast way to lose good candidates who simply do not have that much unpaid time to give strangers. Save the practical test for later in the process, once you already have a shortlist you are genuinely excited about.

They also do not work well for roles where the real job cannot be meaningfully scaled down, or where the only way to test someone properly is to give them access to sensitive systems, real client data, or unsupervised responsibility they have not earned yet. In those cases, a structured interview with scenario questions is usually the safer substitute.

Is it fair and legal to test candidates like this

Rules around trial shifts, unpaid work, and minimum pay vary significantly by country and even by region within a country. Before you build any practical test into your hiring process, check your local labour regulations and official government guidance on trial periods, unpaid trial work, and minimum wage rules. This is not something to guess at.

As a general principle that holds up almost everywhere: if a candidate is producing real work that benefits your business, whether that is serving actual customers, writing content you could publish, or fixing something that was actually broken, they should usually be paid for that time. Short, scoped skills exercises that a candidate completes on their own time, purely to demonstrate ability rather than to produce usable output for you, are more commonly treated as unpaid, but this still depends heavily on local rules.

Keeping the process fair, not just legal

Beyond pay, fairness means consistency. Every candidate applying for the same role at the same stage should get the same task, the same instructions, the same amount of time, and the same tools. The test itself should be clearly relevant to the job, not a random puzzle that has little to do with day-to-day work. And you should be ready to offer reasonable accommodations, such as extra time or an alternative format, for candidates who need them.

Get agreement in writing before the test starts, even if it is a short email. State what the candidate will do, how long it will take, whether they are paid, and what happens to any work they produce.

Designing the test so every candidate has the same shot

Start from the real job, then scope it down to something that fits in a reasonable window. Thirty to ninety minutes is typical for a desk-based exercise. A paid trial shift for a hands-on role might run half a day. Longer than that and you should think hard about whether you are testing skill or just extracting free labour.

Write the brief once and reuse it exactly for every candidate. If it is a take-home task, send identical instructions, deadlines, and reference materials to everyone. If it is a live, in-person exercise, use a script so each candidate gets the same introduction, the same context, and the same level of help from staff.

Avoid trick elements, artificial time pressure, or ambiguous instructions that are not actually part of the job. If the real role never requires someone to work under a countdown timer, do not add one just to see how they cope with stress. Test the skills the job needs, nothing more.

It is worth piloting the test internally first. Ask a current employee to complete it and time how long it genuinely takes. This catches unclear instructions and unrealistic time limits before candidates hit them.

Paying candidates and handling logistics

Decide your payment policy before you invite anyone. Paid trial shifts are the norm for hands-on roles like hospitality, retail, and trades, where the candidate is doing real, billable work alongside your team. Short take-home exercises are more often unpaid, provided they are genuinely brief and do not produce work you intend to use.

Whatever you decide, put a clear cap on it. If a take-home exercise should take sixty minutes, say so, and design it so that is actually achievable. If you are asking for a half-day trial, pay for it, and confirm the rate and timing in writing beforehand.

Sort out equipment and access in advance. If the test needs a laptop, specific software, tools, or safety gear, either provide it or tell the candidate exactly what they need to bring. Reimburse any reasonable expenses, such as travel to an in-person trial.

Running the session on the day

Brief the candidate properly before they start. Nerves distort performance, and a candidate who does not know where to park, who to ask for, or what the first hour will look like is not being tested fairly against one who does.

Assign a friendly point of contact who checks in on them rather than leaving them to sink or swim alone. This matters both for fairness and for how the candidate experiences your business. Even people you do not hire will talk about how they were treated.

Hi [Name], looking forward to having you in on [date]. You'll be working with [contact name] on [task], and it should take around [time]. We'll cover [pay/expenses] as discussed. Just bring [items], and [contact name] will meet you at [location/time]. Let me know if you need anything adjusted.

Keep conditions realistic but not artificially harsh. Do not deliberately withhold information, sabotage equipment, or throw in surprise obstacles to see how someone reacts, unless handling genuine surprises is a real and common part of the job. Give feedback or at least a clear timeline for next steps once the session ends.

Scoring fairly: rubrics and multiple evaluators

Build a simple scoring rubric before you run the test, not after. List the three to five things that actually matter for this task, tied directly to the requirements of the role, and score each on a consistent scale, for example one to five. Score against the rubric, not against your gut feeling.

Where possible, have two people score independently and then compare notes. This catches individual bias and gives you a more reliable picture, especially when candidates are close in ability. Write down the reasoning behind each score, not just the number, so you can explain your decision later if asked.

If you are running practical tests across several candidates for the same role, keeping the scheduling and scoring organised gets harder fast. Tools like Hyrewell can help here: candidates apply through one link, get automatically screened into a ranked shortlist, self-book their practical test or interview slot, and receive an e-signed offer once you have decided, so the whole process stays consistent and easy to compare across everyone who applies.

Common mistakes to avoid

Using the test as an early screening filter for every applicant instead of a later step for your shortlist
Giving different candidates different tasks, time limits, or levels of help
Not paying for real work that benefits the business
Skipping a written rubric and scoring on impression alone
Making the test far longer or harder than the actual job requires
Failing to confirm pay, timing, and expectations in writing beforehand
Leaving candidates unsupported or unbriefed on the day
Not offering accommodations for candidates who need them
Using the exercise to get free work rather than to assess skill

Sample wording you can copy

Use something like this when inviting a candidate to a paid trial or practical test:

Thanks for the great conversation last week. As the next step, we'd like to invite you to a paid practical session on [date], covering [task]. This will take approximately [time] and will be paid at [rate/flat fee]. You'll be working alongside [name/team] and will need [equipment/access]. Please confirm you're available, and let us know if you need any adjustments to take part.

And for a short, unpaid take-home exercise:

As part of our process, we ask shortlisted candidates to complete a short exercise similar to real work in this role. It should take no more than [time] and there's no need to go beyond that. Please send it back by [date]. This is unpaid and used only to assess fit for this role; we won't use the work for any other purpose.

Being upfront in this way, before the candidate has committed any time, is one of the simplest ways to keep the whole process fair and to protect your reputation as an employer.