Joining a startup means your job description is a starting point, not a boundary. In the first month you'll usually do some of what you were hired for, plenty of what nobody planned for, and learn which of the two the company actually needs from you.
What it is
A small company with fluid roles is one where headcount is low enough that work gets assigned to whoever is nearest and capable, not to whoever owns that function on an org chart. The first month is when you find out how far that goes, and it goes further than most people coming from a structured employer expect.
Beyond Tick Boxes: Diversity Recruitment Strategies That Actually Transform UK Workplaces
Master the Virtual Hot Seat: 7 Video Interview Techniques Recruiters Don't Tell You
AI Governance Careers Surge in 2026: Why This Role Will Define the Next Decade
Why it matters now
Teams are staying smaller for longer. Most scaling companies I deal with would rather hire one adaptable engineer than two narrow specialists until the product is settled, and that preference is why you see job ads that read like three jobs stapled together. If you're moving from a large employer, you'll meet this model sooner than you think. The ONS publishes labour market data if you want the wider picture on how smaller firms make up the UK employment mix, but the day-to-day reality is easier to see from inside one.
How the first month actually runs
I'll use engineering because it's what I know. The same shape shows up in marketing, ops and sales at small companies, just with different tools.
Week one
You get access to things. Repositories, the cloud console, Slack or Teams, the project board in Linear or Jira, the shared drive with a folder structure nobody has tidied since the second hire. Someone will send you a README that's six months out of date. Read it anyway, then ask what's wrong with it. That question gets you a better picture of the system than the document does.
In a corporate setting there's usually an onboarding plan with named buddies and a schedule. Here there might be a founder who says "have a look around and shout if you're stuck." That's not neglect. It's genuinely how they work, and it's a test of whether you can orient yourself.
Weeks two and three
A request comes in. A customer can't log in, say. In a larger company that goes to support, then to a ticket queue, then to a team with an on-call rota. Here it lands in a channel and someone writes "anyone free?" That someone is increasingly you.
So you pick it up. You check the logs, find that a token expiry is misconfigured, patch it, and then, because there's no QA function, you test it yourself and write the note to the customer success person explaining what happened. One bug, four hats: developer, tester, incident responder, and the person who writes the post-incident summary nobody asked for.
This is the bit that surprises people. The work isn't harder, it's wider. And the habits that served you well in a structured team (waiting for a clear brief, staying in your lane, escalating upwards) quietly stop working.
Week four
By now you'll have a sense of which hat is yours. Most people find one or two areas where they're clearly stronger than anyone else in the building, and the company leans on that. The other hats you wear because somebody has to. Pay attention to which ones you don't mind. That tells you more about where you'll grow than any appraisal template will.
The tools you'll meet
Expect fewer of them than at a big employer, and used less formally. A typical engineering setup is GitHub or GitLab, a CI pipeline (often GitHub Actions), a cloud provider, Notion or Confluence for docs, and a monitoring tool such as Datadog or Sentry. Product, design and commercial people share many of the same spaces, which is deliberate. There's no separate release management office to hide behind.
Process is lighter too. You might have a weekly planning session and a Friday demo instead of formal sprint ceremonies. That's not sloppiness. At twelve people, a two-hour retrospective costs more than it returns.
Who it suits
It suits people who get energy from seeing a problem and just fixing it. It suits people who can live with a half-written spec and ask the right question to complete it. It suits anyone who wanted to see how a whole business works and has been frustrated by working on one small corner of a large one.
It doesn't suit everyone. If you rely on clear role boundaries to do your best work, or you need predictable hours and a documented escalation path, you'll find the first month tiring. That's no failing. I've watched strong engineers leave big employers for small ones and be back within a year, and I've watched others never look back. Both outcomes are fine. The mistake is not knowing which you are before you sign.
And one thing worth saying plainly: if the founders can't tell you what you'll own in the first three months, that's information. Fluid roles shouldn't mean no direction.
How to get started
Before you accept an offer, ask three things. Who made the last hire and what did they end up doing that wasn't in the advert? What happens when something breaks at 9pm? Who decides priorities when two people want the same hour of your time?
Once you're in, write things down. Your own notes on how the system fits together will be more useful than anything in the wiki, and in a few months you'll be the one improving it. Take the awkward jobs early, because that's how you learn the business. And ask for feedback in week three, not week twelve. Small companies rarely schedule it for you.
If you're weighing up roles, browse current tech and startup jobs to see how employers describe this kind of position, and read the career insights on OHUB for how candidates are approaching the move. Compare the advert wording against what the interviewers say. The gap between the two is where the real job lives.
What would you refuse to do in your first month, and would the company be right to ask?