On this page: Key takeaways · When is a trial shift worth running? · What should a trial shift rating sheet contain? · What can a trial shift tell you, and what can it not? · Does a proefdag have to be paid? · How do you keep a trial shift safe? · What are the most common trial shift mistakes? · What should agencies agree with the client site? · The bottom line · How Radius helps · Frequently asked questions
Key takeaways
- Write three to five observable criteria in advance and give every candidate the same tasks.
- Work samples predict job performance moderately on average, but they are not automatically fair.
- We found no official Dutch guidance on paying for a proefdag; our advice is to pay for productive work and agree terms in writing.
When is a trial shift worth running?
A trial shift is worth it only when three answers are yes.
- Is it a skill needed on day one? Safe reach truck handling by a trained operator qualifies; your scanner system, which you teach in week one, does not.
- Can it be done safely and under supervision? A named supervisor watches throughout, tasks stay within the candidate's training, and your insurer has confirmed that candidates on site are covered.
- Are the criteria written down? You can list, in advance, what you will see when the task is done well.
If any answer is no, use another method. A job knowledge question rated against a model answer, such as "What do you check on a reach truck before your first pallet?", tests what someone knows at a fraction of the cost; see how to ask about the work instead of counting years. The research agrees: work samples generally do not suit entry level hiring, where skills are learned after hire [1].
What should a trial shift rating sheet contain?
The sheet fixes tasks, criteria and record before the trial starts. Write it with the site supervisor and your own safety instructions.
Illustrative example: a practical check for a reach truck operator in a warehouse.
| Field | What to write |
|---|---|
| Requirement tested | Operates a reach truck safely (needed on day one) |
| Checked before the trial | Certificate or training record seen; safety briefing given; adjustments asked about in the invitation |
| Tasks, the same for everyone | 1. Pre use check of the truck. 2. Put away three pallets at height. 3. Retrieve one pallet from a marked location |
| Criteria, written in advance | Completes the pre use check and reports any fault; travels with the load lowered; checks the rack and aisle before lifting; slows down and signals at crossings; places each pallet square and stable |
| Record for each criterion | Met, not met or not observed, with one line on what you saw |
| Supervisor and length | The supervisor's name; 45 to 60 minutes including the briefing |
| Outcome | The decision and its reason, written on the day and signed by the supervisor |
Decide on the criteria alone: "seemed keen" is an impression, not an outcome.
Hiring for warehouse roles? See how Radius supports warehouse and reach truck hiring.
What can a trial shift tell you, and what can it not?
A trial shows what someone can do today, while watched. It cannot show how they will work at 05:30 in week six, with nobody watching. It also separates knowing from doing: a spoken answer shows someone knows the steps, a practical check shows they can carry them out. In the evidence levels of What Is the Signal Gap?, the first is structured evidence and the second demonstrated.
On a scale where 0 means no relationship with later performance, Sackett and colleagues estimate 0.33 for work sample tests on average, against 0.42 for structured interviews, mostly from studies outside the Netherlands [1][2]. Candidates tend to accept practical tests: across 38 samples from 17 countries, work samples and interviews were the methods applicants rated most favourably [4].
Does a proefdag have to be paid?
We found no official Dutch guidance that answers this directly (checked 21 October 2026). These sources come closest:
- The law's definition. The Dutch Civil Code defines an employment contract as an agreement to work in the service of an employer, for pay, for a period of time (Book 7, Article 610) [5]. Whether a particular trial falls under it depends on the facts; this article does not answer that.
- A separate scheme. The UWV trial placement (proefplaatsing) lets an employer try out someone who receives certain UWV benefits, normally for up to two months, while they keep their benefit. It needs UWV's agreement, and the employer must intend to offer a contract of at least six months [6][7].
- A voluntary code. The NVP Recruitment Code says an orientation or shadowing day may not lead to disguised labour with wage value (section 4.3) [8].
- Union guidance. The FNV tells candidates that many employers will not pay for a trial day and advises putting arrangements in writing [9]. That is advice, not law.
Our view, as good practice rather than legal advice: if the candidate does productive work, such as serving real guests or putting away real stock, pay for it. Agree in writing beforehand what the trial involves, how long it lasts, whether it is paid and how travel costs are covered.
How do you keep a trial shift safe?
Treat the candidate like a new starter in their first hour. The Working Conditions Act (Arbowet) also counts as an employer anyone who has another person work under their authority, even without an employment contract (Article 1(2)), and requires employers to inform workers about the work, its risks and the protective measures, and to instruct them (Article 8) [10]. How this applies to a given trial is a legal question, but it is a clear reason to give the full site briefing.
The Working Conditions Decree requires that self propelled mobile work equipment, such as a reach truck, is operated by workers with specific expertise (Article 7.17c) [11]. So a trial is not the way to find out whether someone can drive one at all: check certificates first, then watch how they work, with the supervisor beside them and an agreed stop rule.
What are the most common trial shift mistakes?
Most mistakes come from treating the trial as an informal look.
- Testing trainable skills, or judging without written criteria. Fix: run the three question test and use the rating sheet.
- Different tasks for different candidates. Fix: same tasks, same order, comparable times.
- No adjustments. Dutch law requires effective adjustments for people with a disability or chronic illness, as needed, unless disproportionate, including when a vacancy is filled (WGBH/CZ, Articles 2 and 4(a)) [12]. Fix: describe the tasks in the invitation and ask about adjustments.
- Assuming a practical test is fair. In US applicant samples, the average gap between Black and White candidates on work samples was 0.73 standard deviations, far above the 0.38 often quoted [3]. Fix: tie every task to the job, and review tasks that many otherwise qualified candidates fail.
What should agencies agree with the client site?
Agree who supervises, who insures and who decides, before the trial. The client's safety rules and supervisor apply on site; put in writing who briefs the candidate, pays for any productive work and records the outcome. The NVP code notes that recruitment through agencies falls under the separate Recruiter Code [8].
The bottom line
A trial shift earns its cost when it answers one question nothing else can: can this person safely do a task the job needs on the first shift? Write the criteria first, keep it short and supervised, and settle pay and insurance in writing.
How Radius helps
Radius Hire is the AI hiring platform for frontline roles: applicants take a short, structured voice interview, and the recruiter sees the evidence behind each requirement. A voice interview cannot watch someone drive a reach truck, so practical checks stay with your site. Because the AI Interviewer asks every applicant the same questions, in the language set for the role, you can save trial shifts for where they add something. Where the employer uses Radius Passport, certificates are recorded with their expiry dates, so a recruiter can review them before booking a check. Radius never rejects a candidate on its own; a recruiter reviews the evidence and makes every hiring decision. See how this works for warehouse and hospitality teams.