Skip to main content
Back to guides

Career guide

First 90 Days

You already got the job. Now you have about twelve weeks to prove you can do it well.

By James Bugden, Career Coach. Senior Recruiter @ Uber

40 min read

Based on "The First 90 Days" by Michael D. Watkins.

Sign in to save your progress and access all guides

Where are you right now?

There is a moment in every new job where you stop costing more than you contribute. Watkins calls it the break-even point.

His research puts it at around six months for a typical mid-level hire. It takes longer if you came from outside into a company already doing well.

Ninety days will not get you there. What ninety days gives you is a reputation.

That reputation decides how fast the rest goes.

People decide what they think of a new hire early, often within the first few weeks. Once they decide, what they tell you changes to match.

Think well of you, and they share more. If they think poorly of you, they share less.

Your next decision gets worse. That worse decision confirms what they already thought.

This guide is the plan for that.

Before day one

The transition starts the day you accept the offer, not your first day of work.

The weeks between signing and starting are the only time in the whole job when you have hours to spare and no accountability for anything in your new role. You will never get that combination again. Watkins calls this window the fuzzy front end. He treats it as a serious asset.

Learn the whole company, not only your patch

Read the obvious things: strategy, products, recent results, recent news. Then read the things that govern how you can have impact.

  • How does planning work here, and on what cycle?
  • Who reviews performance, and how?
  • How do people get promoted?

Those three answers tell you more about what is possible in this job than the org chart does.

If your level warrants it, ask for a briefing pack before you start. Ask to visit a site, sit in on a call, or meet two or three people informally. Nearly every company says yes to this and almost nobody asks.

Talk to your predecessor if you possibly can

The person who had the job before you knows things that are not written anywhere. Ask the recruiter or the manager to make the introduction if they left on good terms. If they were pushed out, you likely won't get the meeting. That itself is information.

The question that opens people up is not "what should I do." It is "how did we get to this point." Ask that about the team structure, about the roadmap, about the thing everyone complains about. You will hear the history behind decisions that look irrational from outside. That keeps you from tearing down something built for a reason you have not learned yet.

Close out the old job in writing

Your current manager has an incentive to keep you doing your old job for as long as possible, not out of malice but because you are about to become their resourcing problem.

So negotiate the handover as soon as you have resigned. Agree what you will finish, what you will hand over, and, most importantly, what you will not do. Then write it down and send it back to them in an email. Not as a challenge. As a summary. Doing two jobs at once is one of the most reliable predictors of a rough transition. The worst version is still doing the old job at night while you learn the new one during the day.

Make yourself a breakpoint

Almost nobody gets a clean handover, so make one up. Pick a specific weekend. Treat it as the line. Before it, you were the person who did that job. After it, you are the person who does this one.

This sounds soft. It is the most practical thing in this section. The biggest single cause of a failed first ninety days is not a skill gap. It is continuing to run the old playbook. That playbook worked before. It is what you are good at. Nobody told you to stop.

While you still have the new joiner's advantage, ask about culture

Ask these of your manager, your recruiter, and anyone who will take a coffee with you before you start. The answers show how work gets done here. That is often different from what the handbook says.

  • ●How do things get decided here? Does it come from the top, or from whoever builds the most agreement first?
  • ●What are meetings for? Real debate, or ratifying something already settled in private?
  • ●How does conflict get handled? Openly, avoided, or pushed downward?
  • ●Who gets recognised? The visible driver, or the quiet person the team relies on?
  • ●Are there ways to get results here that would not be acceptable, even if the result was good?

What kind of job did you take

Twenty minutes here changes everything downstream.

Two people start the same week at the same company with the same title. One of them makes fast calls on partial information and gets promoted.

The other makes fast calls on partial information and gets managed out. The difference is not talent.

They walked into two different situations. Only one of them worked out the right way to work in their own situation.

Watkins sorts new-role situations into five types and calls the model STARS. It is the most useful thing in the book.

The right behaviour in one situation is the wrong behaviour in another. Nobody's job description tells you which one you are in.

Check yourself against two things for each situation below.

  • How it feels to be in the room day to day.
  • What your job is once you have named it.

Both change completely from one situation to the next. That is the whole reason the diagnosis matters.

Start-up

You are assembling something that does not exist yet. A new team, a new product line, a new market, a new process.

The mood in the room

Excited confusion

Your real job

Decide what not to do. The energy is already in the room. Focus is the scarce thing.

Turnaround

Already broken, and everybody knows it. You were brought in to stop the bleeding. Nobody needs convincing that something has to change.

The mood in the room

Close to despair

Your real job

Give people a concrete plan and the confidence it will work. Cut back to a defensible core, then build out from it.

Accelerated growth

The thing works and it is scaling faster than the structure around it. Everything is on fire in a good way.

The mood in the room

Undisciplined momentum

Your real job

Get people to accept more process than they want. Do not kill the momentum that made it necessary.

Realignment

It used to work. It is drifting. Nobody senior has said so out loud. The numbers are not bad enough to force the conversation.

The mood in the room

Denial

Your real job

Make the need for change visible before you try to lead it. Evidence first, then direction.

Sustaining success

You inherited something good from someone people liked. The team already knows how to win.

The mood in the room

Complacency

Your real job

Invent the challenge. Play good defence first, then find the next level.

The exercise: split 100 points

You will almost never be in a pure situation.

  1. Split the points. Take everything you are responsible for. Split one hundred points across the five types, based on where your effort should go over the next ninety days. The points have to add up to one hundred. That is the part that does the work. It forces you to say what gets less.
  2. Check your favourite. Then do the second half, which most people skip. Write down which of the five situations you enjoy most. If your top allocation matches your favourite, check whether your preference wrote your priorities, not the other way round. That is how a turnaround specialist ends up smashing up a team that only needed steering.
  3. Keep the result. It is also the clearest way to explain to your manager and your peers what you are doing and why. Explained that way, you will not sound like a new person announcing changes.

The situation nobody wants is the one that pays

He surveyed leaders on which of the five they found hardest, and which they preferred. The two lists are almost exactly inverted.

Watkins's survey of leaders

Realignment: rated hardest by the largest group. Preferred by almost nobody.

Start-up: rated hardest by the smallest group. The runaway favourite.

The reason is not difficulty. It is credit.

A turnaround produces a before and an after, and everyone sees which one you caused.

When a realignment goes well, it produces a crisis that never happens. Nobody calls the power company to thank them for the lights staying on.

The conclusion is uncomfortable. Organisations reward turnarounds lavishly.

They have no idea how to reward realignments. That means nobody gets paid to stop businesses from becoming turnarounds in the first place.

Week 1, the listening tour

The first five days are not about performing. They are about learning how this place works.

You do not have to prove anything in week one. Your job is to collect information you can compare. Do that well. By the end of the month, you can say something true about this place, something the people here have not said to each other.

Run the same five questions with everyone

Meet people one on one. If you manage a team, start with your direct reports. If you do not, start with the people you will work with: your closest teammates, the leads of the teams next to yours, and whoever has been here longest. Same opening about yourself, same questions, same order.

The five questions

  1. 1What are the biggest challenges this team is facing, or about to face?
  2. 2Why is the team facing them?
  3. 3What is the most promising opportunity nobody is working on?
  4. 4What would have to happen for us to go after it?
  5. 5If you were me, what would you focus on first?

The script matters more than any individual question. Identical questions give you answers you can lay side by side. Then you can see where people agree, where they contradict each other, and who is being more open than the rest. Different questions give you five interesting conversations and no pattern.

The script also protects you from the loudest person. Whoever gets to you first, or speaks best, will shape your view of this place more than they should. A fixed script is the cheapest defence against that.

Read how they answer, not only what they say

The second layer of these meetings is about the person, not the content. Do they answer the question you asked, or the one they wanted? When something in their area breaks, do they own it, explain it, or point at another team? Can they describe how their work connects to other teams, or do they only think about their own?

Watch for the topics that produce a strong reaction. Those are the things the person cares about. They show you what kind of change would energise them, and what kind would feel like an attack.

What to do

  • ●Listen more than you talk. Take notes in every meeting, visibly.
  • ●Run the same five questions with every person, in the same order.
  • ●Ask everyone: how did we get to this point?
  • ●Find the docs, the wiki, the decision log, the old chat threads. Read.
  • ●Learn names, faces, and what each person owns.
  • ●Set up your tools and access on day one. Do not wait for IT to chase you.

What not to do

  • ●Do not push for changes in week one. You have not earned the read yet.
  • ●Do not say the old way at your last job was better. Even when it was.
  • ●Do not promise dates before you understand the system or the team.
  • ●Do not skip the small introductions. People remember being skipped.
  • ●Do not vent in private channels. You are still new to everyone in them.

Who you must meet: your manager, every direct teammate, the lead of any next-door team whose work touches yours, and the person who has been here longest. That last one matters most. They know why things are the way they are, and they are almost never on the onboarding list.

What to write down: the words this team uses, who owns what, the unwritten rules of "we do not do that here", and everything that strikes you as odd. Keep the odd list. In two months you will have stopped noticing any of it. Right now, your view is the only fresh perspective the company has.

At the end of the round, put together what you heard: observations, open questions, and a couple of early hypotheses. Then get the group in a room and give it back to them. You will learn a second time from how they argue with it. In about four weeks, you will have shown that you understood the place. That single meeting does more for your standing than any amount of early output.

The five conversations

Your manager sets your bar, explains you to everyone else, and controls what you get.

Most new people take the situation, the expectations and the resources they are handed. By month four, they find out someone else had already set the terms. The alternative is to shape it. Five conversations do the shaping.

These are not five meetings. They are five topics that come up again and again in every conversation you have with your manager for the next three months. What matters is you know which one is active, and you say so at the start.

01

Situational diagnosis

Week 1, then again once you have your own read

Find out how your manager sees the situation you inherited. Which parts are a turnaround, which are drifting, which are fine. How it got this way.

Good looks like: You can say back, in their words, what they think is broken and why. Your view can differ. It cannot surprise them that it differs.

02

Expectations

First week, no exceptions, then reset at day 30

What does good look like in the short term and the medium term? How will your performance be measured, and by when?

Good looks like: Two questions, answered in writing: what has to be true at 90 days, and who else, besides your manager, gets a say in judging it.

03

Style

Week 1 to week 2, revisited the first time something rubs

How you two will work. Preferred channel, frequency, and the shape of your decision box: what you decide and report, what you consult on first, what stays theirs.

Good looks like: At first, expect a small box of decisions you can make alone. It should get bigger. If it does not, that is the conversation.

04

Resources

From week 3, once you can say what you need and why

A negotiation, not a request. Money and headcount, but also air cover, access, a name introduced, a meeting you need to be in.

Good looks like: You presented a menu, not a number. This outcome costs this much. That outcome costs that much. Pick.

05

Personal development

Around day 90, not before

Where are you strong, and where do you need to grow? What might you take on that develops you without wrecking your focus?

Good looks like: It happens at all. Most people never ask, then wonder why nobody sponsored them.

Ambiguity does not default to neutral

If you and your manager remember the agreement differently in six months, their version wins, not yours. That is not unfair. This is how authority works. It is the whole reason to make expectations explicit and write them down.

So ask the same question more than one way. What does good look like at ninety days? Would anything make you think this hire went badly? Who else, besides you, has a view on whether I am doing well? That last one catches the finance director or the regional lead whose opinion counts and who nobody mentioned.

The five don'ts

  • ●Do not stay away. If they do not reach out, you do. Being left alone can feel like trust. It is often only distance.
  • ●Do not surprise them. Late bad news is a much bigger mistake than bad news. The worst version is them hearing it from someone else.
  • ●Do not bring only problems. Do not wait for a full solution either. Bring the problem plus how you plan to start on it.
  • ●Do not run down your checklist. Bring at most three things you need them for.
  • ●Do not expect them to change. Adapting is your job, not theirs.

Things that work

  • ●Pick three things your manager cares about and mention progress on them every time you meet.
  • ●Ask for resources as a menu. This outcome costs this much. That outcome costs that much.
  • ●Underpromise. Getting a lot done still counts as a miss if you promised more.
  • ●Find out what is untouchable. Some products, people and processes are personal to your manager.
  • ●Get good marks from the people your manager already trusts. Their view of you is partly hearsay.

The 30, 60, 90 day milestones

Three gates, three deliverables, three meetings you book yourself.

After a couple of weeks in the job, write a ninety-day plan. It does not have to be long. Bullet points are fine. It has to be written. The point is not planning. It is agreement.

Share it with your manager and treat it as a contract for your attention. The plan says what you will do. More usefully, it says what you will not do. That sentence protects you in month two, when three new requests arrive. Then split it into three blocks with a review meeting at the end of each. You book those meetings. Nobody else will.

Phase 1

Day 30

Show a diagnosis

What you owe: A read on the situation, the priorities you have picked, and a written plan for the next 30 days.

  • ✓Ran the same five questions with every person who matters to your work
  • ✓Can name the STARS mix of what you inherited, and say why
  • ✓Have written expectations from your manager, in their words
  • ✓Removed at least one small, persistent irritant nobody else fixed
  • ✓Booked the review meeting and sent the plan before it, not during it
Phase 2

Day 60

Show progress and a commitment

What you owe: Progress against the plan you agreed at 30, and what you are committing to deliver by 90.

  • ✓Launched two or three early-win projects, not seven
  • ✓Fed your findings back to the team and let them argue with you
  • ✓Mapped who decides the things you need decided
  • ✓Asked for the resources you need, as a menu with prices
  • ✓Started the weekly log. Date, what changed, who was in it
Phase 3

Day 90

Show a result and a direction

What you owe: One result the team can point at, and a conversation about where you go next.

  • ✓Delivered something visible that was on the agreed list
  • ✓Held a real review, not a chat, and asked for the score
  • ✓Opened the personal development conversation yourself
  • ✓Proposed one change with the evidence attached
  • ✓Set the next quarter's goals and said them out loud to your manager

The first block is for learning, and you have to negotiate it

Unless something is on fire, the first thirty days belong to learning and to building personal credibility. That is a deliberate allocation, not a soft start. It only works if you agree it out loud with your manager, and then hold them to it.

Inside the block, run a weekly rhythm. Ten minutes at the end of each week: what did I say I would do, what happened, what is next week for. It sounds trivial, but it is the difference between arriving at the thirty-day review with a diagnosis and arriving with a feeling.

What the three review meetings are for

The day-30 review

You bring a diagnosis of the situation, the priorities you have picked from it, and a written plan for the next thirty days. The plan should say where you intend to look for early wins. The conversation draws on two threads: situational diagnosis and expectations. What you want out of it is agreement on what kind of situation this is, a clear bar, and a yes to the next block.

The day-60 review

Progress against what you agreed at thirty, and a specific commitment for what will be true at ninety. If you manage people, this is where you give your first honest read on the team. Otherwise, it is where you ask for the resources your early wins need. By now, you can say exactly why.

The day-90 review

One thing the team can point at, and then the personal development conversation. Where am I strong, where am I not, and what should I take on next. Ask for it explicitly. Almost nobody does. The people who do are the ones managers start describing as ready for more.

Building credibility early

Four questions people are silently answering about you in your first two weeks.

People already know something about you before you show up. Someone here has talked to someone who worked with you, read your profile, or drawn a conclusion from where you came from. You start with a reputation, whether it is accurate or not. From then on, people will notice the evidence that fits it.

Decide, deliberately, whether you will confirm the story people expect or quietly contradict it. Both are valid. Doing neither leaves the first impression in charge.

What everyone is working out about you, without asking

  • ●Can this person make a hard call and hold it?
  • ●Do they have values I recognise and would want to work near?
  • ●Do they have the right kind of energy, or the exhausting kind?
  • ●Do they expect a high standard, from themselves as well as from me?

Six balances, and both ends are a failure

Each of these is a pair, and the mistake is always going too far one way. Most people know which end they lean towards. Few have said it out loud.

1

Demanding, but possible to satisfy

Hold people to what they committed to. If you are never satisfied, people stop trying to satisfy you. Know when to say that was good.

2

Accessible, but not too familiar

Be easy to approach, but stay able to make an unpopular call. This is harder if you were promoted over former peers.

3

Decisive, but judicious

Make a few low-consequence decisions fast so people see you can decide. Openly park the ones you are not ready for, and say when you will be.

4

Focused, but flexible

Be clear about what matters. Let people get there their own way when the route is theirs to choose.

5

Active, without causing commotion

Make things happen without making everyone around you sprint. Watch the stress you are generating, not only the output.

6

Willing to make hard calls, and humane about it

People watch what you do. They watch how you do it more closely. A fair process survives an unpopular outcome. An unfair one does not survive a popular one.

Fix the small annoying things fast

Every team has two or three irritants that everybody complains about and nobody owns. A meeting that should be an email, or one that runs an hour and needs twenty minutes. A broken doc, a dead link, an access request that takes four days. A supplier nobody has spoken to properly in a year.

Pick one in your first fortnight and remove it. This is not a strategy. It is the cheapest credibility available. You are visibly spending your newness on something that helps other people, not on something that helps your review.

Learning in public is an action, not a delay

The urge to do something big in your first three weeks feels like pressure from the organisation. It rarely is. That urge comes from inside you. It is usually a confidence problem, not a business one.

Being visibly committed to understanding the place is protective. This makes it hard for anyone to say you turned up with the answer already written. But you also have to look like a quick study. You have to know when to stop asking and start deciding. Learning past about week six starts to read as avoidance.

Calibrating with your manager: ask in week two, "At the end of 90 days, how will you know I am doing well?" That question forces your manager to name the real bar, not the polite one. Write the answer down in their words. Read it again at day 60 and check what you are spending your time on.

Early wins that count

"Get a quick win" is the most repeated and least useful piece of new-job advice out there.

The instinct is to find the easiest visible improvement and bank it. Watkins calls this the low-hanging fruit trap. His image for it is exact: you have built a rocket that is all first stage. It gets off the ground and then comes back down, because there was nothing behind the initial thrust.

The correction is uncomfortable. Sometimes you have to leave the easy fruit alone and reach higher up the tree. An early win has to do two jobs at once: move something that matters, and build the way of working you want more of.

Score each candidate out of 16

List the things you might go after. Score each of them from 0 to 4 on these four questions. Then read the total, and override it with the rule below.

1

Does this make a real difference to how the team performs?

Score this one first. A zero here kills the idea, no matter how well it scores on the rest.

2

Can you do it quickly, with what you already have?

Inside your first 90 days, with people who already work here and budget that already exists.

3

Does it build toward the goals you and your manager agreed on?

If it is not on the list you wrote down in the expectations conversation, it is a hobby.

4

Does the way you would do it model the behaviour you want more of?

A win delivered in the wrong way costs more than it earns. Delivered in the right way, it is worth two.

The rule that overrides the total: a zero on the first question kills the idea, no matter how many fours it collects on the other three. An easy, fast, on-strategy, behaviour-modelling project that does not change how the team performs is a well-organised waste of your best month.

How you win is part of the win

A result achieved by going around people, or by leaning on your newness, or in a way that cuts against how this place expects things to be done, will cost you more than it earns. The same result, achieved in a way that shows what you want more of, counts twice. You have shipped the outcome and the example at the same time.

What counts as a win also depends on the situation you diagnosed earlier. In a realignment, getting people to talk straight about the state of things is a genuine achievement. In a turnaround, the same activity is a month you did not have.

Change comes in waves, and the pause is deliberate

The pattern that works is not one long push. It comes in stages.

  1. A period of focused learning.
  2. A first wave of changes.
  3. A deliberate slowdown, while things settle and you learn what happened.
  4. A deeper wave, with much better information.

The pause is the part people skip. If you never stop changing things, you cannot tell which change worked.

And the people around you never get a moment where the ground is stable. The first wave is only there to buy you the credibility and the relationships that make the second wave possible.

When to push back on scope

Saying yes to everything in month one is the most common self-inflicted wound in this whole guide.

New people say yes because they want to look useful. Refusing anything also feels risky before you have any standing. Then the work does not all get finished. The thing you were trying to protect is the thing you lose.

Pushing back does not make you difficult. It shows your manager the real arithmetic while there is still time to change it. In week three, it reads as judgement. By week eleven, after a miss, it reads as an excuse.

Signals it is time to speak up

  • ⚠The scope grew but the deadline did not move.
  • ⚠Someone asks you to own something three layers deep in your second week.
  • ⚠Two leaders pull you in two directions. Neither knows about the other.
  • ⚠You are the only person on a project that needs a team.
  • ⚠The ask landed in chat, never in writing, and keeps changing shape.

Words you can borrow

Scope grew, deadline did not move

I want to hit this and hit it well.

To do that, I need to drop X or move the date by a week.

Which one do you want?

Too many priorities at once

I have A, B and C on my plate this week.

If I do all three, none of them will be good.

Which one matters most to you?

Ask is fuzzy

Can you tell me what done looks like here?

I want to make sure I build the right thing the first time.

A rough example of a good outcome would help.

Never say yes in the room

The most useful habit here is to defer commitment by default. "That sounds interesting. Let me think about it and come back to you today." If you are pushed for an answer immediately: "If you need an answer right now it has to be no. If it can wait until this afternoon I will give it a proper look."

Start from no. It is easy to move to yes later, and it costs you nothing. Moving from yes to no is a retraction, and people remember those. Notice that requests for future commitment are easier to agree to. The reason is your calendar three months out looks empty. It will not be empty when you get there.

Undefended boundaries get tested until they move

If you never say what you will not do, other people will decide for you. They will keep going until something breaks. It gets worse over time, and that is the unpleasant part. The more you take on without complaint, the less people value it, and the more arrives.

You cannot ask other people to protect boundaries that you have never drawn. Draw one early, on something small, and hold it. People will test it once. After that, they tend to respect it.

Getting things done without authority

Authority gets less useful as you rise, not more.

This is the surprising one. Everyone assumes seniority makes things easier to push through. In practice, positional authority becomes less decisive the higher you go, for two structural reasons. The problems get more ambiguous, so you have fewer answers you can prove with data alone. And everyone else in the room was also selected for ability and drive, so they push back harder.

So everything you learned about getting things done by being right and being in charge stops scaling at almost the exact moment you get promoted for being right and being in charge. If you are an individual contributor, this section matters most to you. You already have none of the authority and all of the dependency.

Map who decides

Take one thing you need to get done. Write down who has to say yes, what you need them to do, and by when. Then write down the group who might block it. You are not looking for unanimity. You need a critical mass on each side of the deal. Some people have too much invested in the current arrangement to change their mind. Trying anyway is how three months disappear.

A second structure runs underneath the org chart: who people go to for advice, who is believed, who has been here long enough to know things no one else remembers. Watkins calls it the shadow organisation. You find it by watching. In meetings, notice who defers to whom on the hard questions, and where people's eyes go when a heated topic comes up.

The five things that make someone influential here

  • ●Expertise. They are the person who knows.
  • ●Information. They see things before other people do.
  • ●Connections. They know everyone, in every direction.
  • ●Resources. They control budget, headcount, or access to something scarce.
  • ●Loyalty. Someone senior trusts them personally.

None of these is a job title. The person with the most of them in your area is the person to have a coffee with in week two.

Why people say no, and what to do about each one

Before labelling anyone an obstacle, work out what they are protecting. There are five common reasons, and the counter is different for each. That is why "they are simply difficult" tells you nothing useful.

They are comfortable with how things are

Change threatens a position or a relationship they like. Find out which, then work out what they get in the new arrangement.

They are afraid of looking incompetent

They think they will be bad at the new way. Offer the training, the pairing, or the runway before you ask for the yes.

It cuts against something they believe

They think you are rewarding the wrong behaviour. Argue the principle, not the plan. You will not win this one with a spreadsheet.

It costs them power

You are moving a decision away from them. Say so out loud. Hidden power grabs get punished later, at the worst possible moment.

It hurts people they feel responsible for

This is the most honourable reason to oppose something. It is also the one most often mistaken for politics. Address the harm and you usually win the person over.

One useful piece of social psychology sits behind all five. When people behave in ways we do not like, we overestimate how much of it is personality. We underestimate how much of it is the situation they are standing in. The person blocking you might be carrying a goal, an incentive, or a promise you cannot see. Ask what would have to be true for them to say yes.

Five techniques that do the actual work

Ask, then say it back

Consultation is not politeness. It is persuasion. Ask real questions, listen to the answer, then summarise what you heard. The summary is where the influence happens. You choose the frame there.

Frame it for the person in front of you

Three registers. The data. The principle. The reason to care. Different people move on different ones. Most arguments fail because they carry only one.

Go a step at a time

People will walk a path they would never agree to jump. Start with a meeting to define the problem. Then the options. Then the criteria. By the time you get to the decision, it often makes itself.

Choose the order

Who you talk to first decides how hard the rest is. Get the respected sceptic on side early and arm them with your best argument. They will make the case in rooms you are not in.

Create something that forces a date

A review, a demo, a standing meeting, a deadline someone external cares about. When nothing forces action, one person's delay becomes everyone's excuse.

Documenting your wins

Memory lies. A log does not.

Once a week, ten minutes, same slot. Write down what changed because of you, when it happened, who else was in it, and the number if there is one. Also write what you tried that did not work. That is the material for the development conversation at ninety days. You will not remember it either.

This is the same weekly review the ninety-day plan runs on. You are not keeping two systems. You are keeping one, and taking two different things out of it: next week's focus, and a permanent record.

Four things the log is for

  • ●Your probation review, which will be written from your manager's memory unless you supply better material.
  • ●Your first performance cycle, which will cover months you have forgotten by then.
  • ●Your promotion case, which needs specifics with dates attached and cannot be reconstructed later.
  • ●Your next resume, and the interview after it, where the difference between a good answer and a great one is one number.

One habit that pays for itself: when someone thanks you in writing, paste it into the log with the date. Not for your ego. Because it is evidence in someone else's words, and in a promotion case that is worth more than anything you write about yourself.

Probation period notes

Rules vary a lot by country and by company. Start with what your contract says.

Taiwan, common case

Probation is usually 3 months. The law does not set a length. But 3 months is the market norm, and 6 is not unusual at senior levels.

Labour insurance, health insurance and the pension contribution all still apply during probation. Severance and notice follow the Labour Standards Act.

If the company wants to end probation, they still need a real reason. They cannot say "not a fit" and be done.

United States

Most states are at will, which means either side can end the job at any time. Probation is a company policy, not a legal status.

60 to 90 day introductory periods are common. Some benefits, health insurance in particular, might only start after that window. Read the fine print.

This matters more if you hold a work visa such as H-1B or L-1. If you are let go, you usually have 60 days to find another job.

United Kingdom

Probation is contractual, not statutory. 3 to 6 months is the norm, and contracts often set a shorter notice period during it.

Protection against ordinary unfair dismissal generally requires two years of service, so the practical bar for ending a new contract is low.

Discrimination protection applies from day one and does not depend on service length. So does statutory notice after the first month.

Continental Europe

Probation length is usually capped by law or by a collective agreement. The caps differ by country and by seniority.

Notice periods during probation are typically short but real, and they often step up the moment probation ends.

Where a collective agreement covers your role it can override what your individual contract says. Ask HR which one applies to you.

Wherever you are, the practical advice is the same. Find out the exact end date of your probation in week one. Book a conversation for two weeks before it, not after it. And make sure the person who signs it off has seen written evidence of what you did. In most companies, that decision gets made fast, by someone working from memory.

This is general guidance, not legal advice. Read your contract or ask a lawyer before signing.

Top 5 early stage mistakes

Other people made these mistakes. You do not have to.

Watkins lists seven transition traps that the whole book is built to prevent. These are the five that do the most damage in the first ninety days, in the order I see them most often.

  1. 1

    Running the playbook that got you hired

    The trap: Sticking with what you know, and arriving with the answer

    You were hired partly for what you did somewhere else. That is not the same as being hired to do it again here. The fastest way to lose a room is to describe the fix in week two. Doing that tells everyone you did not understand the problem. It also shows you were not listening when they explained it. Wait until you can explain why the current thing exists before you propose replacing it.

  2. 2

    Doing something only to be seen doing something

    The trap: The action imperative

    The pressure to act in your first month rarely comes from your manager. It comes from you. What looks like a productivity problem is a confidence problem. Visible learning is an action. Being the person who asked the question nobody had asked is a result. Nobody has been let go at day 45 for learning carefully.

  3. 3

    Never agreeing what success means

    The trap: Setting unrealistic expectations

    You can perform well and still fail, if you were measured against a bar nobody wrote down. This is the single most common way a good hire ends up with a bad probation review. Get the bar in writing, from the person who sets it, in the first week. Then check it again at day 30, because it moves.

  4. 4

    Starting six things

    The trap: Attempting to do too much

    Six projects at twenty percent each get you six things that are eighty percent unfinished when your review lands. Two or three priorities is the whole budget. Everything else is a note in a doc with a date on it.

  5. 5

    Learning only the technical layer

    The trap: Wrong type of learning, and neglecting peers

    Most new people learn the systems and skip the culture and the politics. Then they get surprised by a decision they never saw coming. Most people also over-invest down and up, and under-invest sideways. Your peers are the ones who will decide, in the corridor, whether you get help when you need it.

When it is not going well

The dip at three to six months is normal. Almost nobody warns you about it.

Somewhere between month three and month six, most people in a new role go into a valley. The early energy is gone. So is the novelty. What is left is the actual size of the problem. It is common enough that Watkins treats it as a stage, not a warning sign. Keep working the plan and you come out the other side of it.

Knowing it is coming is most of the defence. People who do not expect it conclude they made a terrible decision. They start updating their CV during the exact fortnight the job was about to get easier.

Four ways this goes wrong, and what each one looks like

Undefended boundaries

Looks like: You never said what you would not do. People kept asking, and you kept saying yes.

What helps: The more you give here, the less you are respected and the more you are asked for. Say no once, early, on something small. Then hold it.

Brittleness

Looks like: You called something early, and it is not working. Backing off now feels like admitting you were wrong.

What helps: It gets more expensive every week you wait. Reframe it as new information rather than a reversal, and move.

Isolation

Looks like: You are hearing only the official version. People have stopped bringing you bad news, or they never started.

What helps: Find the people who do the work. Ask about problems specifically, and thank the first person who brings you one.

Work avoidance

Looks like: There is one hard call you keep not making, and you are busy with other things.

What helps: Name it in writing, put a date on it, and tell someone the date. Hard calls do not decay. They compound.

Take stock properly, in writing

Set aside half an hour and answer these truthfully. Write them down, not in your head. In your head, the exercise turns into a mood.

  • ●How do you feel? Excited, confident, in control? For every no, ask why not, and what you might do about it this week.
  • ●Who have you failed to connect with, and what is the real reason?
  • ●Which meeting has bothered you most since you started, and what specifically bothered you?
  • ●Which decisions have turned out well, and which badly, and why?
  • ●What opportunity did you miss, and was it blocked by you or by something outside you?

The last question is the test. Experienced people are as likely as anyone else to put every difficulty outside themselves. The cost of doing that is not guilt. It is passivity. Fine, if that is the situation. If part of it is you, that is the part you can change this week.

Build the three advisers you need

Most people have one kind of adviser and think they have a network. You need three kinds, and the mix changes as you get more senior. Early in a career, the mix is mostly technical. Later, political counsel decides outcomes.

Technical advisers

People who can tell you whether the thing is good. Usually inside the function, sometimes outside the company.

Cultural interpreters

People who can tell you what happened in that meeting and why nobody said anything. Usually long-tenured, rarely senior.

Political counsellors

People who can tell you how a decision gets made and who has to be in the room. This is the one most people never build.

Keep some of them outside the company. Insiders tell you how things work here. Outsiders tell you whether "how things work here" is reasonable. They are loyal to you, not your employer. That is different, and sometimes essential.

Quick reference cheat sheet

Tick as you go. Saved to this device.

Before you start

Week 1

By day 30

By day 60

By day 90

Your action plan

Do these this week. In order.

Resources

Where this came from, and what to read next.

The source

"The First 90 Days: Proven Strategies for Getting Up to Speed Faster and Smarter" by Michael D. Watkins.

The frameworks named in this guide are his: the break-even point, the STARS model, the five conversations, the low-hanging fruit trap, the seven transition traps. The survey figures quoted are from his research. The recruiter observations are mine, and the book does not contain them.

This guide is based on "The First 90 Days" by Michael D. Watkins. Recruiter perspectives are from James Bugden's 15+ years of hiring experience.

Did you find this guide helpful?

Free

Get full access, for free

Create your free account to unlock every guide, tool, and template on this site.

  • 10+ career & interview guides
  • Resume Builder & AI Analyzer
  • Job research & salary tools
Create a free account

No credit card required

More from this stage

  • Guide

    How to Resign

    Leave your old role cleanly. Scripts, timing, and the resignation letter template.

    Read guide →

Share

Found this guide helpful?

Share it with a friend who's job hunting. Good content deserves to be passed on.

Keep reading free

Create your free account to read the rest of this guide and save your progress.

Already have an account?

SOAR
© 2026 SOAR. All rights reserved.Privacy PolicyHelpTerms of Service