On Friday you were really good at your job. On Monday there’s a message from your boss with a party emoji in it, a new title, and six names under “direct reports”. Somebody takes you for a coffee and says they’re proud of you. Nobody says what you’re supposed to do on Tuesday.
That’s the usual deal. Most people step into management with no training for it at all.
of people stepping into UK management have had no formal management or leadership training. CMI, 2023.
Gartner found that 40% of managers in their first two years are struggling to support their team. So if you’re quietly making it up as you go along, you’re in the majority. You’re also not doomed.
The first 90 days are more predictable than they feel. A handful of small habits, started early, do most of the work. The rest of this playbook is those habits, week by week, with the exact words to use. But first, a quick trip to Scranton.
The Michael Scott problem
Michael Scott, regional manager
Michael Scott runs a paper company’s Scranton branch in the American sitcom The Office. He’s a salesman who ended up in charge, and the joke that keeps the show going is that nobody ever worked out what the new job actually was. He wants to be liked more than he wants the branch to run well. The mug on his desk says “World’s Best Boss”, which tells you whose approval he’s after.
He’s fiction, obviously. But anyone who’s been promoted without being told what the new job is will recognise the pattern: keep doing what got you promoted, and judge yourself by whether people like you.
The real-world cousin is less funny. One large study of sales firms found that the best salespeople were the most likely to be promoted into management, and that once promoted they did worse as managers than the firms expected. The firms had leaned on past sales results when choosing. What predicted a good manager better was how much the person had worked jointly with colleagues, and that was under-weighted.
Nobody tested why the top sellers struggled. Our guess, and it is only a guess, is that the old job is where you know you’re good, so it’s where you drift back to.
Here’s a quick way to check yourself. Open your calendar for last week. If it’s mostly your old work with a few meetings on top, you haven’t started managing yet. You’re a very senior team member with a worse diary. The next 90 days are about shifting that balance, a bit at a time.
Weeks 1 and 2: listen, and keep your hands in your pockets
Book 30 minutes with every person in your team. Don’t pitch a vision. Nobody has asked for one, and you don’t know enough yet. Ask four things:
- What’s working well here that I should never change?
- What’s draining you at the moment?
- What would help you do your best work?
- What do you need from me?
Write down what you hear and change nothing. That’s harder than it sounds. You’ll hear three things you could fix by Friday. Leave them. People are watching what you do with what they tell you, and the first thing they need to learn is that you can be trusted with it.
I’m not changing anything yet. I want to see how it really works first. By the end of the month I’ll tell you what I’ve done with what you’ve told me, even if the answer is “nothing, and here’s why.”
Three more jobs for these two weeks.
1. Ask your own manager what good looks like for you. Not “any advice?” Ask for three outcomes at day 90, and write them down while they say them. If they go vague, send the three you think they mean in an email and ask them to correct it. That counts as getting it in writing.
If we sat down on day 90 and you were pleased, what would have happened?
2. Agree three outcomes with each person in your team. Only 46% of employees clearly know what’s expected of them at work (Gallup, 2025). Be the reason your team is in the 46%.
3. Stop promising things. “I’ll try to sort that” is how new managers pick up promises they can’t keep, and people remember every one. When people believe a promise has been broken, trust and satisfaction go down and plans to quit go up (Zhao et al., 2007). Keep a private list of everything you’ve said you’ll do. Clear it once a month, or say out loud that you can’t.
Weeks 3 to 6: set the rhythm
Three habits, in this order, because each one makes the next one easier.
1. A fixed one-to-one with each person. Same slot, every week or fortnight. Their list first, one blocker you’ll remove, one commitment each. Move it if you have to. Never cancel it, because every cancelled one-to-one tells the person where they rank. (We wrote a whole playbook on this: How to run a one-to-one that isn’t a status update.)
Sorry, something’s come up. Can we skip this week?
I can’t do 2pm. Can we do 4pm today, or first thing tomorrow?
2. A Friday reset, 20 minutes. Write down the three outcomes that would make next week a good one. Give estimates as ranges, not single dates, because people predict they’ll finish sooner than they do. Pick the one task you’ll be tempted to dodge and decide when and where you’ll start it. Then tell your team what isn’t happening.
If only three things happen next week, it’s these. Everything else gets a good-enough version.
3. A proper team launch. Even a team that’s been running for years deserves 90 minutes where you answer six questions together: who’s in, why we exist, what good looks like, who decides what, how we work, and when we check. How well a team does depends heavily on conditions the leader sets up: a real team, a clear direction, a structure that helps and well-timed coaching (Hackman, 2002). So you run the session, and you talk the least. Come back to it at 30 and 90 days.
The Starting Grid
The six-box canvas for launching or relaunching a team. Ninety minutes, reviewed at 30 and 90 days. It sits alongside TRIP, the Friday reset, in the Ducora OS.
Weeks 7 to 12: the first hard bit
By now you know the team well enough to do the things you couldn’t do in week one. These are the ones that separate people who manage from people who merely have the title.
Give your first piece of real feedback, early. Within 48 hours of the moment, about the work rather than the person. That window is Ducora practice, not a tested finding. The point is not to save it for the review. Nobody can fix what they don’t know about, and the review turns into an ambush. (More in How to give feedback that changes what people do.)
Deal with the thing everyone’s waiting for you to deal with. Almost every inherited team has one: a lovely colleague whose work has been slipping for months. While it’s left, teammates carry the load, then start to withdraw (Felps, Mitchell and Byington, 2006), and your strongest people notice you haven’t acted. You’re allowed to say you should have raised it sooner. It beats saying nothing.
I should have said this three weeks ago, and I didn’t. I want to talk about the last three deadlines.
Ask for feedback about you. One person, privately, about one thing: “What’s one thing I could do differently in Monday’s meeting?” Thank them, whatever they say, and show what you changed within two weeks. If the first honest answer gets defensiveness, it’ll be the last.
Count something. A change launched with energy fades when nobody keeps score. In a big test of gym-visit programmes, 45% worked while they were running, and only 8% were still working four weeks after they stopped (Milkman et al., 2021). That’s gym members, not teams, so treat it as a warning rather than a law. Pick one new behaviour, count it weekly for ten weeks, and put the number where the team can see it.
Your 90 days on one page
The plan above, boiled down to a single sheet you can stick next to your monitor and tick off.
- The week-by-week plan, with the scripts
- The ten checks to tick off by day 90
- Space for your three day-90 outcomes
Five ways it goes wrong (and the quick fix)
“It’s quicker if I do it myself.” It is, today. It also makes you the bottleneck and stops your team growing. Hand over one thing a month, and say out loud who decides. (How to delegate without it all coming back to you)
“Let me look at that for you.” Fixing someone’s work in your evening feels kind. They learn that bad drafts get quietly repaired, and you learn to dread Sundays.
“I’ll save it for the review.” See above. Feedback that arrives six months late isn’t feedback, it’s a charge sheet.
“Yes, we can do that by the 30th.” Answer with a trade-off instead: “Yes, and here’s what drops.”
“They’re good people. They’ll sort themselves out.” Good individuals don’t automatically become a good team. Launch it.
What if you’ve been in the job a while?
Everything above still works. You don’t need to wait for a new start date. Pick your own day one, tell the team it’s a reset, and run weeks 1 and 2 as if you’d just arrived. All it costs you is one slightly awkward sentence: “I’d like to do this better, and I’d like your help.”
What the evidence says, and how sure we are
- StrongManagers matter for who stays. In one large firm, managers rated better at people management had lower team turnover, and lifting a manager’s people skills from the 10th to the 90th percentile predicts around 60% lower turnover (Hoffman and Tadelis, 2021). Whether one-to-ones are the route isn’t tested.
- ModerateTop individual performers promoted to manager did worse as managers in sales firms, and collaboration history predicted managerial success better than past sales (Benson, Li and Shue, 2019). One large dataset in sales. Holding on to the old job is one plausible reason, not a tested one.
- ModerateTeam effectiveness depends mostly on conditions the leader puts in place (Hackman, 2002). Influential, but not tested as a single model.
- StrongWhen people believe promises have been broken, mistrust goes up and satisfaction goes down (Zhao et al., 2007).
- ModerateA single negative team member can trigger defensive reactions in the rest of the team (Felps, Mitchell and Byington, 2006). Moderate as theory, Emerging for direct tests.
- EmergingProgrammes that fade after they stop (Milkman et al., 2021) come from gym visits, so they’re an analogy for teams, not a finding about them.
- Honest gapNobody has tested this exact 90-day sequence. It’s a Ducora plan built from findings on each part: one-to-ones, clear expectations, team launch, early feedback. The parts are evidenced. The wrapper is ours.
Frequently asked questions
What should I do in my first week as a new manager?
Listen. Book a 30-minute conversation with each person in your team, ask what’s working, what’s draining them and what they need from you, and write down what you hear. Ask your own manager what good looks like for you at day 90. Resist the urge to change anything yet.
How often should a new manager hold one-to-ones?
Weekly while you’re new, fortnightly once things settle, and monthly at the very least. In one small study, one-to-ones only helped engagement when they happened at least monthly, and their quality mattered more than how many you held. That’s early evidence, so treat monthly as a floor rather than a law.
What if my own manager gives me no direction?
Ask three questions and write down the answers: what does good look like for me in 90 days, what are the two or three things you most want to see change, and how would you like me to keep you updated? Then send the answers back in an email and ask them to correct anything you got wrong.
How long until I feel like a manager?
Nobody has a reliable number. In studies of new habits, routines took about two months on average to settle, with a very wide range, so give yourself a full quarter before you judge how it’s going.
Should I be friends with the team I used to be part of?
You can be friendly. What changes is that you now make decisions that affect their pay, their workload and their reputation, and they know it. Tell them plainly that things will feel a bit different, and that you’d rather say so than pretend they won’t.
Sources
- CMI (2023), on UK managers’ training.
- Gartner (2023), on managers in their first two years.
- Gallup (2025), on employees who know what’s expected of them.
- Benson, Li and Shue (2019), on promoting top performers to manager.
- Hackman (2002), on the conditions for team effectiveness.
- Zhao, Wayne, Glibkowski and Bravo (2007), Personnel Psychology, on broken promises.
- Hoffman and Tadelis (2021), Journal of Political Economy, on manager people skills and turnover.
- Milkman et al. (2021), on programmes that fade after they end (gym visits).
- Felps, Mitchell and Byington (2006), on how a negative team member affects the rest of the team.
