The Working Interview, Run From Trial Invite to an Offer Decision

You got invited to a working interview, trial shift, or multi-day paid work trial, and now you need to run it well from the moment you accept the terms through the final decision. This guide is the operator's sequence for that: classify the trial, lock scope and pay in writing, check your wage rights, perform on the day, and convert the result into an offer or a clean walk-away. It is for candidates in any field where someone asks you to do the real job before they decide, from a restaurant floor to a startup codebase.

Most public writing on working interviews either defines the term or argues about its legality. Neither tells you what to send before you accept, what to do in the first hour, or how to get paid when it ends. This fills that gap.

What a working interview is, and when it must be paid

A working interview is a trial where you perform real tasks in the real environment instead of answering questions about them. It must be paid when your output stops being an evaluation and starts being work the business benefits from.

The legal spine is the FLSA's "suffer or permit" standard combined with the "primary beneficiary" test: time spent predominantly for the employer's benefit is work, and time spent predominantly for your own benefit is not, judged across the totality of the circumstances. A written aptitude exam, a typing check, or a physical agility screen almost never has to be paid. A trial shift where you ring up customers, cook food that gets served, or stock shelves alongside regular staff almost always does. Employers must pay for all compensable work they know or have reason to know is being performed, even work they did not expressly assign.

The single most important thing to understand is that the label is legally irrelevant. Calling a shift a "tryout," a "shadow day," or an "interview" does not change the analysis when you are doing the job. Your strongest lever as a candidate is documenting whether your output was actually used.

I did not find a single DOL enforcement action captioned specifically as a "working interview" wage case in the public record, so I will not claim one exists. The doctrine above is well sourced; named enforcement examples are not. If you need to establish exposure for a specific situation, check your local wage authority's hours-worked guidance and, where the stakes justify it, an employment lawyer in your jurisdiction.

Where you are changes whether the trial can be unpaid

Jurisdiction, not industry, sets the pay floor. The same trial can be legal unpaid in one place and illegal in another, which is why classifying where you are comes before everything else.

Ontario's Working for Workers Four Act (Bill 149) amended the definition of "employee" effective March 21, 2024, confirming that work performed during a trial period counts as training, which is compensable work. In New York, under state labor law and the FLSA, employers must compensate candidates for any work that benefits the employer or serves actual customers, even during a trial shift or working interview. Asking an applicant to work an unpaid trial that serves customers is against the law there.

The UK is the grey zone. There is no outright ban. Trial shifts sit under the National Minimum Wage Act 1998 and the 2015 regulations, with government guidance last updated in July 2022. A short, supervised trial used only to assess ability can be unpaid; once you do productive work or the trial runs long, you count as a worker who must be paid at least the National Minimum Wage, which was at least £11.44 per hour for those aged 21 and over as of April 2024. The government's position is blunt: an unpaid trial lasting longer than one day is highly likely to require payment in all but very exceptional circumstances. The teeth behind this are real. HMRC can order arrears, add a penalty of up to 200% capped at £20,000 per worker, name the business publicly, and back pay is recoverable for up to two years.

What named employers actually run

Real programs tell you what to expect and what a fair structure looks like. Durations run from a single paid day to five, and the credible ones pay a daily rate agreed in advance.

Linear runs a paid 2 to 5 day period where you work with the team on a real project you plan to implement, with access to internal tools, paid at a daily rate communicated in advance. SageOx runs a paid 3 to 5 day in-person trial in the real codebase, with terms agreed in writing before it starts and pay for every day whatever the outcome. PostHog uses a one-day SuperDay with representative rather than shipped work. Automattic assigns a roughly 40-hour project paid at approximately $25 per hour, which drew criticism as below market. Restaurant trials generally run 3 to 6 hours depending on the operation and level.

Two things matter here. First, the "representative versus real" distinction is a term you must pin down: Linear and Gumroad ship real work, PostHog deliberately does not. That difference decides who owns your output. Second, a prestigious name is not a proxy for a fair rate. Benchmark the daily rate against a pro-rated slice of the role's real salary.

The terms to lock before you accept

Before you accept, get seven things agreed in writing: scope, supervisor, success criteria, pay rate, pay processing and timing, duration, and work-product ownership. A verbal promise is not a term; it is a hope.

There is no single canonical candidate checklist in the public record, so treat this as the synthesized spine rather than a cited list. SageOx documents the employer-side norm you should mirror: terms agreed with you in writing before the trial starts, at a rate appropriate to the role, no contact with your current employer, and pay and confidentiality agreed in writing beforehand. General pre-offer practice adds the rest: knowing what success looks like and how it is measured, who you report to, and how you will be paid.

Here is the message to send once you have the invite. Refusing to start until this is answered is the behavior of a professional, not a problem.

Thanks for the trial invite. Before I confirm a date, I want us aligned in writing on a few things so the day runs cleanly:

1. Scope: what specific work will I do, and what should exist at the end?
2. Supervisor: who directs my work and who evaluates it?
3. Success criteria: what does a strong result look like, and how is it measured?
4. Pay: what is the rate, and is it paid for the full trial regardless of the outcome?
5. Processing: how and when am I paid, and what do you need from me to run it?
6. Duration: start time, end time, and total days.
7. Work product: who owns what I produce, and is it used or kept as a sample?

Happy to confirm the date as soon as these are set. Looking forward to it.

If you are applying across many postings and want your materials, scope questions, and fit framing consistent from one trial invite to the next, Refolk writes your resume from your own history, tailors it to each posting, and scores how well you actually fit, which keeps the pre-trial conversation grounded in evidence rather than guesswork.

Run the day, in order

This is the full sequence from invite to paid outcome. Each step has an owner and a definition of done so you can execute it without improvising.

On the day itself, the hospitality pattern and the tech pattern converge on the same shape. A good host prepares before you arrive, welcomes you, introduces you to the whole team, then assigns specific duties with everything you need and never tells you to float. They watch the clock and tell you when you are nearing the end. The tech version is scope agreed at the onset, then work in the actual tools, standups and reviews, and things that ship. If the employer does not run the kickoff well, run it yourself: ask for scope, the success bar, and who to ask when stuck.

How this goes wrong

Working interviews fail candidates in seven recognizable ways. Each has a false positive that looks fine on the surface and a specific check that exposes it.

The selectivity signal deserves extra weight because it is the cleanest single diagnostic. Pay should correlate with how few candidates reach the trial. If almost every applicant gets a work trial, you are probably looking at mass free labor dressed as a final round, not a selective last step. So "how many people reach this stage?" is a real question, not small talk, and the answer tells you whether to proceed.

The label does not change the analysis when you are doing the job; your output is the evidence.

Judge the odds, and judge them back

A well-run paid trial can raise your conversion odds, not lower them, because you demonstrate directly instead of through proxies. Treat the trial as a two-way audition and evaluate the employer as hard as they evaluate you.

The numbers support optimism about the format when the terms are fair. Linear reports over 50 people scaled with a 96% retention rate across four years of work trials, and its CEO has said the post-trial rejection rate is under 50%. A Sequoia-published trial week showed a hire rate around 66% out of the trial. Those are far better odds than most multi-round interview loops produce. The mechanism is simple: real work removes the guesswork on both sides.

Use the close-out hour to collect feedback and form your own verdict. Ask what they saw, what they would want more of, and when you will hear back. By the end you should have made a judgement on the potential employer, not only awaited theirs. If the day was disorganized, if you were told to float, or if pay stayed vague, those are data about how they will manage you after you join.

The model is spreading, so expect it more often

The trial model is migrating from thin-margin hospitality into tech, which is a much larger labor pool, and that raises the wage-law stakes because tech trials carry daily rates and shippable output.

In Refolk's index of professional profiles, the US holds about 7,967 tech-recruiting professionals against roughly 459 in the UK, and about 624 US hospitality-sector recruiters, with Marriott International the single largest employer in that sample. The ratios tell the story: US tech recruiters outnumber UK tech recruiters about 17.4x and US hospitality recruiters about 12.8x, both derived from those counts. A practice born in restaurants and clinics is now designed and run by a far larger population of tech recruiters, which is exactly why this stage is worth a standard of its own.

If you want to understand who runs these programs before you walk in, Refolk can surface them by name and context from public records.

Verify before you call the job done

Run this checklist before, during, and after the trial. If any item is unchecked, you have an open risk to close before you accept, start, or walk away.

What to do next

Keep this current by re-checking two things each time you get a trial invite: your jurisdiction's current wage floor and guidance date, and whether the specific work will be used. Those two inputs drive every decision in this guide, and both change. The procedure above does not.

The habit that protects you most is the smallest one: never start a trial on a verbal pay promise, and always keep the written terms. That single discipline converts a grey-area request into a clean transaction, whether it ends in an offer or a walk-away. Everything else in this playbook is built on top of that record trail.