Trial days and work trials

What is a work trial?

You turn up, almost always in person, and do the work you would do as a full-time hire.

Not a puzzle, not a hypothetical, not a case study. A real project on the team's roadmap, with access to their tools and their people. Length varies a lot, from a single day to meet everyone up to two weeks, and occasionally a full month.

Some companies call it a work trial, some a trial day, some invent their own name for it. Linear runs a paid two-to-five-day version and puts every hire through it, executives included. The mechanics are broadly the same wherever you meet it.

Why do startups do this instead of more interviews?

Because an interview shows them your personality, and they need to see how you work.

A technical interview tells a founder whether you can solve a problem they set. It doesn't tell them how you operate across a normal day, how you handle a question with no clean answer, how you talk to the people you'd sit next to, or whether the team enjoys having you around.

In the first ten hires that matters more than anything on your CV. Execution beats the idea every time, and execution comes down to who is in the room. So the founder wants the most accurate read they can get, and watching you do the job is the most accurate read available.

Where does the trial come in the process?

Almost always last, after the interviews and right before an offer.

It costs the company real money and several people's time, so nobody runs a trial on a candidate they aren't already sold on. If you've been invited, you've passed. That should change how you walk in.

One exception worth knowing. A founder may ask early, in a first or second interview, whether you'd be willing to do a trial at all. That's a commitment test, not a scheduling question. Saying no there can end the process on the spot.

How long do trials last, and what does the length tell you?

Anywhere from one day to a month. Longer trials tend to come from more established startups.

A Belgian cybersecurity company flew a friend of mine in from Madrid for a single day. He spent it meeting the seven-person team and sitting in on a sales call with the CEO, where his job was to take notes and pull out everything useful. He didn't do any selling himself. At the other end, month long trials exist and they usually come from companies with something to protect (YC-backed, repeat founders, real funding) because only those companies can afford a month of someone's salary and attention on a maybe.

That's a pattern worth knowing rather than a rule. But if a company proposes a month, they're probably serious, and the commitment they're asking for is proportionally bigger. You'll likely need to leave or pause your current job to do it.

Is this a real trial, or am I being used for free work?

Almost certainly a real trial. The free work theory doesn't survive contact with how startups actually operate.

Think about the incentives. A startup's whole problem is finding people who can execute. If you come in and do good work they want to hire you, because that's the outcome they're paying for. Staging a fake hiring process is an expensive and complicated way to get a few days of work out of one person.

The worry isn't crazy though, and a few things are worth checking before you commit:

• Can you find the founders? Real history, real profiles, previous companies.
• Have they raised anything, or been written about anywhere?
• Have you spoken to each founder on a call, ideally met one in person?
• Is the brief a real project on their roadmap, or something vague and disposable?

If you're still uneasy, ask for a shorter trial. Two days instead of two weeks costs you little and tells you most of what you need to know.

Why you should ask to meet every founder

Because one founder is a third of the company, and you need to know whether you fit with all of them.

A good call with one founder tells you that you get on with one person. Founders aren't interchangeable. They're different people with different temperaments, and at this size the culture is simply the three of them. Speak to one and you're making a decision about your working life on a third of the evidence.

Asking does real work in the other direction too. It shows you're taking the decision seriously, that you're eager to move forward, and that you're not waiting to be led through the process. In early-stage hiring you'll rarely reach a trial, let alone an offer, without having spoken to all of them anyway.

What if they haven't raised anything?

Then you're probably not looking at a job. You're looking at a cofounder or founding engineer situation.

A company with no money can't pay you, so what's on the table is equity and a title rather than a salary. That can be a great deal, but it's a different decision with a different risk profile and it needs a different level of diligence. Have the founders built something before? Is there a working product? Are there customers? Is there revenue? Ask for specifics and expect real answers.

At pre-seed, having no institutional round yet is normal and not a warning sign on its own. No money, no product, no customers and no track record is.

Do you get paid? Who covers the flights?

Good trials are paid. Flights should be covered, and it's fair to ask about both directly.

There's no standard day rate. It depends on the city, the role and the stage, and you're not a consultant with a rate card. Some companies pay for travel but not for the days themselves. Some pay for everything. Some put you up in a founder's spare room, which sounds informal and usually means the calls went well.

On flights, a company that has raised a pre-seed or seed round is sitting on somewhere between a few hundred thousand and several million. They can afford to fly you in. If they ask you to cover it yourself, that tells you something. Not necessarily disqualifying, but worth weighing.

There's no harm in asking plainly. "Is there compensation for the trial, and do you cover travel?" is a normal question and a founder won't think less of you for it.

Should you accept an unpaid trial?

If it's short, local and you've vetted the company, often yes. If it's long, far away, or you can't absorb the lost income, ask to be paid.

A day or two of unpaid work at a company you're excited about is a cheap education even if you don't get the job. You'll learn how they build, meet the team, and find out whether you actually want this.

Two weeks unpaid is a different proposition. If you have savings and nobody depending on you, take the five-year view. Nobody who ended up somewhere great looks back and resents the fortnight it took to get there. If rent is due and there's no buffer, that logic doesn't hold and you're entitled to say so. A funded startup asking for two unpaid weeks can pay for them.

How do you do this while you already have a job?

Carefully, and usually without telling your current employer.

Telling them is a real risk. If the trial doesn't convert, you've announced you're leaving to a company that now knows it.

• Holiday. The most common route in Europe, where leave allowances make a one-or-two-week trial genuinely possible.
• The gap between jobs. If you're already moving, negotiate a start date that leaves a month clear and run your trials in it. The safest version is to sign something first, then trial.
• Weekends. Some startups run six-day weeks and can fit a weekend trial in. Most credible ones will want more days than that, so treat this as a fallback.
• Quit first. Only if you can go a few months without income.

Trying to trial in the evenings after a full day at your own job doesn't work. Don't propose it.

If they ask for a month and you can't give one, negotiate rather than decline. Two weeks is often enough. Telling them you can show in two weeks what they're hoping to see in four sometimes lands.

What should you ask before accepting a trial?

Ask what you'll be measured on, and who currently does the work.

The second and third questions are the ones that change your trial. Knowing what good looks like before you start is the difference between five productive days and five days of guessing. And if someone is already doing the job, an hour with them at the start shows you how the company really works and where the gaps are.

If nobody is doing the work yet and you'd be inventing the role, assume nobody will tell you what to do. Build your own plan and send a short daily recap: what you did, what's next, what you need. Founders want visibility without meetings.

What do candidates get wrong on a trial day?

Drifting through the first days without asking for direction, then performing work to hide it.

This is the failure mode and it's almost always the same shape. You arrive without much context, you're not sure where to start, you don't want to look lost, so you don't ask. Day two passes. By day three you're busy-looking and producing nothing, and everyone can tell.

A founder would much rather see someone who does very little on day one, asks a lot of questions, and is shipping by day five. That curve reads as someone learning fast. Flat output across the whole trial reads as someone who can't.

So on day one, say it out loud. I'm not sure where to start, what would be most useful to tackle first, what would make your life easier. It costs you nothing on day one. It costs you the job on day three.

The opposite mistake is asking about everything, every button and every small decision. The line is roughly this: ask about direction, context and priorities, and decide the small stuff yourself.

What are they actually measuring?

Output, judgement, and whether the team wants you back on Monday.

Output is the most legible part and it varies by function. In go-to-market it's countable: demos booked, impressions, content shipped. Close something during a trial and you're in an extremely strong position. In product and engineering the visible artefact is features shipped, but that isn't the whole score.

Judgement is what the CTO or CPO is really watching. Do you think before you build, or do you ship whatever seemed like a good idea? Do you scope properly? Do you bring user feedback into what you're building, or just your own taste? This comes out in conversation as much as in the work. A founder mentions A/B testing in passing and either you've run one and can say how you'd set it up, or you haven't.

Then there's the quiet part. Everyone you worked alongside gets asked afterwards what it was like to work with you. Have the coffees. Spend real time with the people you'd sit next to.

Running the check-in and the final presentation

Lead it, keep it concrete, and show things rather than describe them.

Most trials include a mid-point check-in and a closing presentation, and both are usually candidate-led on purpose. They want to see whether you can own a room and course correct in public.

The shape is simple. What you've done, shown visually wherever you can. What you'd do next. Where you want their input. In product and engineering that means the features, the tests you ran and the decisions behind them. In go-to-market it's the numbers you moved and how.

What should you be evaluating about the company?

The culture, and specifically how founders behave towards everyone who isn't a founder.

A trial gives you something no interview process does: days inside the company, watching how it really runs. Most candidates spend that whole time being evaluated and forget to evaluate back.

Do you like these people? Would you want dinner with them? That's not a soft question. It's the whole reason to take a startup job over a comfortable one, and if the answer is no the equity won't fix it.

Then watch the gap between founders and employees. Some companies have a clear line between the people who own the company and the people who work there, and that line runs downward through everything. It's visible within a few days if you're looking for it.

Walk away if you see anything illegal, or if it becomes clear you're filling a gap rather than being assessed.