Your First 90 Days at a Startup: A Field Guide for New Hires, Not Founders

The 30-60-90 day plan they teach you in business school will get you fired in your first 90 days at a startup.

I've spent 20 years inside startups. I've watched smart people get hired, ramp up like they were joining a Fortune 500, and get quietly pushed out by day 80. I've also watched mediocre people stick around far longer than they had any right to, because they figured out the real evaluation system before their manager ever sat them down for a structured 1:1.

This is the field guide I wish someone had handed me. It is not for founders. There is no shortage of founder content for your first 90 days at a startup. This is for the engineer, the designer, the product manager, the operations lead, the customer success rep, the marketer who just signed an offer and is staring down a Slack workspace with 200 channels, three half-finished onboarding docs, and no plan.

None of what follows works if you are not doing real work. Visibility does not rescue bad output. It exposes it. The tactics here assume you ship. If you are not shipping, the first thing to fix is shipping. Then come back.

The short version. Your first 90 days at a startup are not a corporate ramp. They are an audition nobody announced. Days 1 to 30, survive without becoming invisible. Days 31 to 60, find one piece of leverage and ship something visible. Days 61 to 90, prove you are essential and set your trajectory. Most people lose this audition by treating it like an onboarding.

Why is startup onboarding so broken?

Startup onboarding is broken because most startups do not have onboarding. The founders are too busy, the manager is drowning, and the docs that exist were written by someone who already left. The whole system runs on pattern matching from people who get it. Wait to be onboarded and you run out of credibility before you run out of confusion.

The "30-60-90 day plan" framework everyone refers to comes from Michael Watkins's book The First 90 Days. The book is excellent. It has sold over a million copies and is required reading at most MBA programs. It was also written for executives stepping into established companies with HR functions, structured handoffs, and a manager who has actually thought about what success looks like.

That is not your situation.

At a startup, your manager is probably running their own delivery in addition to managing you. The HR function might be one operations person who is also handling AP, recruiting, and the Slack outage. The "onboarding plan" you were promised is a Notion page that was last updated nine months ago by someone whose name nobody remembers. The default expectation is that you will figure it out, and the people around you will judge whether you are figuring it out faster or slower than the last person they hired.

This is not necessarily a hostile environment. It is just an unstructured one. The people who succeed at startups are the ones who treat this as the opportunity it actually is. Less structure means more room. More room means more leverage, sooner. The price of that leverage is that you have to build the structure for yourself.

The default expectation at a startup is that you will figure it out. The people around you will judge whether you are figuring it out faster or slower than the last person they hired.

You are also walking into something more specific than corporate onboarding usually accounts for. Most of the first three weeks at a startup is just absorbing the chaos. The phase before this one, your first three weeks specifically, has its own brutal logic. The first 90 days build on that foundation. If you skip the foundation, the rest of this post will not save you.

What should you do in your first 30 days at a startup?

In your first 30 days at a startup, you are doing three things at once. You are absorbing as much context as possible without going dark. You are building credibility through small, visible actions that prove you are oriented. And you are quietly mapping the real power structure of the company, which almost never matches the org chart.

The brutal first-week reality

Your first week is not for output. It is for triangulation. You are figuring out who actually makes decisions, what the founders care about right now, where the landmines are, and which Slack channels are signal versus noise.

Every senior person at the company is asking themselves a single question about you, sometimes consciously, often not: does this person get it.

That question is being answered before you write your first line of code, design your first mockup, or send your first customer email. It is being answered by how you ask questions in your first standup. By whether you read the existing code or design files before suggesting changes. By whether you say "let me look at this and come back" or "we should just rebuild this."

If you want a deeper read on what your first three weeks should actually look like, including how startups secretly evaluate you in week one, the first three weeks at a startup post is the foundation for what follows.

How to ramp without permission

At most startups, "ramping up" is a euphemism for "ask people for things until you can stand on your own." The problem is that the people you would ask are the same people delivering the product. Every question you ask costs them attention they need somewhere else. The new hires who win this trade are the ones who maximize the value per question.

The cheapest thing you can do in week two is read everything that has already been written. Notion docs. Old design files. The pull request history. The customer interview notes. The strategy deck. The pitch deck. The investor update from last quarter. Most of what you would ask has already been answered, and you can compress weeks of asking into days of reading.

When you do ask, ask in writing, in public, and with context. The shape of a good question is always the same.

The good-question template.

1. Context. What you already read or tried.

2. Hypothesis. What you think the answer might be.

3. Specific ask. A binary or short-answer question, not an open one.

Example: "I read the architecture doc. We use X for Y. Is the reason we don't use Z because of A, or because of B?"

That sentence proves you read the doc, you have a hypothesis, and you are not asking the senior person to do all the work. It also creates a searchable record that the next new hire can find. You are doing the company a favor.

What to write down in your first month

Keep a private document with three sections. Update it weekly. Do not share it.

Your first-month notebook.

1. Things I don't understand yet. Acronyms, customer segments, internal tools, why a decision was made the way it was. This list shrinks as you ramp.

2. Things I would change if I had authority. Workflows, docs, processes, missing tools. This list becomes your roadmap once you have credibility.

3. How decisions actually get made. Who in which Slack channel. Which two people quietly agree before a meeting. Which founder overrides which. This is your map of the real company.

This document is the single most valuable artifact you will produce in your first 90 days. By day 90 you will have forgotten what was confusing on day 5, and you will have lost the ability to see the company through fresh eyes. That fresh-eyes perspective is the only thing you have that long-tenured employees do not.

The signals that mean you're already in trouble

Five patterns from 20 years of watching new hires fail. If two or more of these are true by week three, do not wait for the formal feedback that may not come.

1. Your manager stops scheduling 1:1s after week two. That is not them giving you space. It is them losing interest.

2. You are sent to fewer meetings, not more, by week three. The company is deciding whether you are worth the bandwidth.

3. A senior person who was warm in interviews has gone quiet. Find a reason to ask them a specific question in a channel where they have to engage. The act of answering pulls them back in.

4. You are getting feedback that is suspiciously generic. "You're doing great" with no specifics usually means nobody has been paying close attention, which is its own problem.

5. Other people on your team are quietly cleaning up your work. If a peer keeps "tweaking" what you ship before it goes anywhere, they have lost faith in it and have not told you.

The biggest tell is silence. At a startup, attention is the scarcest resource. If you are getting it, you are valued. If you are not, you are being phased out, often without anyone making a conscious decision to do that.

How do you build credibility in your second month at a startup?

In month two, your job is to find one piece of leverage and ship something visible against it. Not five things. One. The new hires who survive their first 90 days at a startup pick a single high-leverage problem in week four, ship a credible first attempt by week eight, and let everyone else watch them work.

How to spot the right first project (and refuse the wrong one)

The wrong first project is the one your manager hands you because they want it off their plate. It is rarely the one that will make you look essential.

The three-test first-project filter.

1. Visible before-and-after. You can point at a number, a screen, a dashboard, or a customer outcome that changed because of what you shipped.

2. A founder or senior leader cares about it this quarter. Not in theory. Not "next year." Right now.

3. Two to three weeks to a credible first cut. If it is bigger than that, you will not finish it before your trust account closes.

If a project fails any of the three tests, it is not your real first project. It might still be useful work. But your real first project is the one you go find yourself. Look at what the founders mentioned in the last all-hands. Look at the metric on the company dashboard that has been red for six weeks. Look at the customer complaint that keeps surfacing in support. That is your real first project.

Managing up when your manager isn't managing

A founder once told me, in a moment of accidental honesty, that "the new hires who succeed here are the ones who manage me, not the ones who wait for me to manage them." They did not mean it as a compliment. They meant it as a description of reality. At a tight-runway company, your manager does not have time to chase you for status updates. They will form a judgment of you based on whether you make their job easier or harder.

Managing up when your manager isn't managing is its own discipline, and it is the soft skill that quietly determines whether you survive your first 90 days at a startup. The single highest-leverage move is a written weekly update.

The Friday update template. Send your manager a short message every Friday. Same shape, every week. Do this without being asked.

• Shipped this week: one to three bullets.

• In progress next week: one to three bullets.

• Risks I'm watching: one thing that could blow up, plus what you're doing about it.

• Need from you: a decision, an intro, an unblock, or "nothing this week."

A manager who gets this message every Friday does not have to ask what you are doing. That alone puts you in the top 20% of new hires at most startups.

How to make your work visible without being annoying

There is a real difference between visibility that builds credibility and visibility that erodes it. The credibility-building kind is short, factual, and lets the work speak. The credibility-eroding kind is performative, takes up airtime in meetings, and signals insecurity. The line between the two is whether the visible thing is the work or you.

Good visibility versus bad visibility.

• Good (the work): A 60-second Loom demoing what shipped. Bad (you): A long Slack post about how hard you worked on it.

• Good (the work): A two-line summary of what you learned from a customer call. Bad (you): A long post tagging six senior people about your "thoughts" on it.

• Good (the work): A pull request description that explains the trade-offs. Bad (you): A standup update that lists every meeting you attended.

• Good (the work): Posting the metric that moved, with one sentence on why. Bad (you): Reposting the metric that moved with five emojis and three exclamation points.

The good kind compounds. The bad kind makes people quietly reduce the meetings they invite you to.

Building your "internal credibility account"

Think of credibility at a startup as a bank account. Every time you ship something on time, you make a deposit. Every time you flag a real risk early, deposit. Every time you make your manager's job easier, deposit. Every time you over-promise and under-deliver, withdrawal. Every time you complain about a thing you have not tried to fix, withdrawal. Every time you bring drama to a senior person without a proposed solution, withdrawal.

By day 60, you should have a balance. By day 90, that balance is what gets you the next project, the next scope expansion, the next conversation about equity refresh. People with balance can spend it on things that matter. People without balance get the cleanup work and the meetings that no one else wants.

Visibility does not rescue bad work. It exposes it.

How do startups actually evaluate you at 90 days?

Startups evaluate you at 90 days through a quiet polling of everyone you have worked with. Your manager's opinion matters, but it is not the deciding factor. The founders ask senior people what they think of you. They ask the engineers if you are a multiplier or a tax. They watch what happens in Slack when your name comes up.

The 90-day moment of truth

At many startups, day 90 marks the end of a formal probation period, and there is often a check-in or review on the calendar. Treat that meeting as the wrong moment. The decision that matters has usually been made before that meeting starts. By the time someone schedules a formal 90-day review with you, they have already polled around. The meeting is the announcement, not the deliberation.

The deliberation happens in a 1:1 between two people you do not see, two or three weeks earlier, based on signals you have been broadcasting since day one. Did this person ramp faster or slower than the last hire at the same level. Did they bring up risks early or hide them until they exploded. Are senior people asking for them to be on more projects, or quietly avoiding pulling them in. The First Round Review framing of "trust account" applies almost exactly to this window. By the time the formal review hits the calendar, the senior people have already decided whether your trust balance is positive or negative.

What this means tactically. Do not save your case for the 90-day review. Make it visible across days 60 to 75, when the polling is happening. The review itself should feel like a confirmation, not a defense.

How to negotiate scope, ownership, and direction

If your trust balance is positive at day 90, you have leverage you did not have on day 1. Use it. Most people leave this leverage on the table because it feels presumptuous to ask for more after only three months. That is a corporate instinct. At a startup, the company is recalibrating its expectations of you constantly. If you do not anchor what you should own next, your manager will anchor it for you, usually below your actual capability.

The day-75 conversation script. Schedule this with your manager. You ask for the meeting, you set the agenda, you bring the document.

"I want 30 minutes to talk about what I should own next quarter.

Here's what I think I've learned in my first 90 days.

Here's the area where I see the biggest leverage.

Here's what I would do if I owned it.

What are you seeing that I'm not?"

That conversation, scheduled by you, is worth more than any performance review your company will run.

When to push for promotion, scope, or equity

Most startup employees never ask for an equity refresh, scope expansion, or title change because they assume those things happen automatically. They do not. They happen when you ask for them and have the receipts to back the ask. By day 90, you should know whether you are going to ask. If yes, prepare for the conversation between day 75 and day 90. Do not have it on day 90. The conversation should be scheduled, written, and specific. "Here is what I have shipped. Here is the value I think it created. Here is what I want next."

You will not get everything you ask for. You will get more than people who do not ask. According to Gallup research on onboarding, employees who feel their early contributions are recognized are far more likely to stay engaged over the long term. The conversation itself is what creates that recognition. Silence creates the opposite.

The questions to ask yourself by day 90

Three questions, honestly answered. Write the answers down somewhere only you will see.

1. Am I doing work I am proud of, or work I am tolerating?

2. Do the people I respect at this company also respect my work?

3. If I had to interview for my own role today, would I be a strong candidate?

If you cannot answer yes to at least two of these, you have a real decision to make in the next 30 days. Pretending you do not is the most expensive mistake you can make at this stage of a startup career.

Your manager is not going to evaluate you. The whole company is.

How do you know if your first 90 days is working?

You know your first 90 days is working when senior people pull you into more, not fewer, conversations, when you have shipped at least one visible thing, and when your manager treats you as a thought partner instead of a task list. If none of those three is true by day 75, spend the next two weeks fixing it.

A simple scorecard. Use it once at day 30, once at day 60, once at day 90.

1. Day 30 hit. I can name three problems the founders care about right now, without checking a doc. I have been added to at least one channel I was not in on day 1. A senior person has said something specific and positive about my work, even casually.

2. Day 60 hit. I have shipped one visible thing. My manager has stopped checking my work line by line. At least one peer has asked for my input on something they are working on. I am running my own 1:1 agenda, my manager is not running it for me.

3. Day 90 hit. I have a clear opinion about what I should own next quarter, and a written plan for it. The founders or my skip-level know my name and what I work on. I have made at least one decision that other people now refer to as "the way we do it." My trust account is in the green.

If you are missing more than one item from any phase, that is a signal, not a verdict. Adjust the next 30 days accordingly. Do not wait for the next checkpoint.

What mistakes should you avoid in your first 90 days at a startup?

The biggest mistakes new hires make in their first 90 days at a startup are waiting to be told what to do, optimizing for looking busy instead of shipping, hiding bad news, and treating the company like a corporate job that happens to be smaller. Each one is fatal alone. All four together is a guaranteed exit.

A list, in rough order of how often I have watched them happen.

1. Asking for permission you do not need. Most of what you are nervous about doing without approval, you can just do. The founders are not going to fire you for trying something reasonable. They will quietly judge you for needing approval to try something reasonable.

2. Confusing activity with progress. Sending 40 Slack messages a day is not the same as shipping. The people watching you can tell the difference. They are tracking what you ship, not how loud you are.

3. Hiding problems until they explode. A small problem flagged early is a deposit in your credibility account. A big problem revealed late is a withdrawal you cannot afford in your first 90 days.

4. Treating it like corporate. Waiting for a 30-day review. Waiting for a structured onboarding. Waiting for HR to set up your 1:1s. None of that is coming. The startup vs corporate mindset shift is exactly this gap.

5. Outrunning your credibility. Pushing for big scope, equity refreshes, or title changes before you have shipped anything. The ask is not the problem. The ask without the receipts is.

6. Not using AI properly. Most of your peers are now using AI as an invisible co-worker on the work that used to take days. If you are not, you will be ramping at corporate speed in a startup that is moving at AI speed.

7. Going dark for a week. If you disappear from public Slack for five business days in your first 90 days, senior people start asking what you are working on. If your answer is not visible, they form a judgment. Not your favorite kind of judgment.

FAQs

What is the 30-60-90 day plan for a startup employee?

A startup-employee version of the 30-60-90 day plan looks different from the corporate version. Days 1 to 30, your job is to absorb context, build a private map of how the company actually works, and make small visible deposits in your credibility account. Days 31 to 60, you find one high-leverage problem, scope it, and ship a credible first attempt. Days 61 to 90, you prove your work is durable, get explicit about what you want to own next, and run the conversation about your trajectory yourself.

How is this different at a startup versus corporate?

At a corporate job, your first 90 days are an onboarding. At a startup, they are an audition. The corporate version assumes you will be told what success looks like. The startup version assumes you will figure it out. The startup vs corporate gap is the single biggest reason great corporate employees fail in their first 90 days at a startup. It is not skill. It is the assumption that someone else is driving.

What if my manager has not given me any goals?

This is the default at a startup, not the exception. Do not wait for goals. Write them yourself. Send your manager a written document at the end of your first week that says, "Here is what I think my first 90 days should look like, here are the three things I think I should ship, and here is how I will measure myself. Tell me what I have wrong." That document is going to be more useful to your manager than anything they would have written for you, and it changes the relationship from them managing you to you managing the relationship.

What if the company is in chaos or about to lay off?

Chaos is the baseline at a startup, not a signal. Real layoff risk is a different conversation. If you have specific signals (board pressure, missed metrics, a sudden hiring freeze, a senior person quietly leaving), the playbook for surviving a startup layoff starts before the layoff happens, and your first 90 days are when you are most exposed because you have the smallest credibility account. The fix is the same as the rest of this post. Ship something visible. Make yourself a multiplier, not a tax.

When should I quit if it's not working?

Not at day 30. Probably not at day 60. The first 60 days at a startup feel terrible for almost everyone, and the people who quit at day 45 usually quit the next thing at day 45 too. By day 90, you have enough information to make a real decision. If your trust account is in the red, your manager is not investing in you, and you cannot name a single thing that has gotten better since week three, the answer is probably to leave. The mistake is not leaving at the wrong time. The mistake is leaving without having actually tried the moves in this post.

If this is your moment to operate intentionally inside the chaos, the ARC framework for startup employees is the underlying structure. Attitude, relationships, competence. The first 90 days are where ARC either takes hold or quietly does not. Each of those three pillars compounds in different ways across the timeline in this post.

For one-on-one help getting through your first 90 days at a startup, coaching with Greg is built for exactly this stage. The pattern of what goes wrong is consistent enough that an outside perspective in week three saves you from spending months figuring it out alone. The book, Enter Startup, is the long-form version of everything in this post.

Your first 90 days at a startup are not your trial. They are your foundation. Build it like the next five years are going to sit on top of it, because they probably are.