How do you design an onboarding that makes new people stay?
Think back to someone's first week in a new job. There is the laptop to set up, the policies to read, the compliance modules clicking by, the org chart no one can quite explain. And then, somewhere in that week, a small thing happens that turns out to matter more than all of it: their manager sits down with them, one to one, for half an hour, just to talk. Not to assess, not to instruct, simply to make contact. That brief, human conversation does more to determine whether the new person thrives, and whether they are still there in a year, than the entire formal programme around it.
This is what Microsoft found when it studied its own onboarding, and the picture of it, the most powerful part of inducting someone being a short, ordinary conversation, makes the argument before a word is said. It is a clean example of designing for belonging rather than assuming it will sort itself out. And the mechanism behind it decides far more than whether new hires stay; it shapes whether anyone, dropped into a new group, ever truly settles.
What Microsoft actually did
Microsoft, like most large organisations, had built onboarding around information and process: the paperwork, the systems access, the training to complete. When the company analysed what actually predicted whether new employees became productive and stayed, the picture that emerged put the social side at the centre.
Two findings stood out. First, new hires whose managers had a simple one-to-one with them in that first week showed markedly better outcomes across a range of measures than those who did not. Second, in a buddy programme, new starters who were paired with an experienced colleague and met them early reported becoming productive faster; among those who met their buddy more than once in the first ninety days, a clear majority said it helped them get up to speed.[1] The thing that made the difference was not a better module or a thicker handbook. It was early, deliberate human contact, designed in rather than left to luck.
The belonging was never going to sort itself out
Microsoft did not try to make onboarding work by improving the information. It built the human contact in deliberately, and let retention follow. That is the reversal worth pausing on. The instinctive approach to onboarding is to optimise the formal: clearer documents, smoother systems, fuller training. We have all been through inductions that were thorough on process and left us, in the part that actually mattered, completely alone, because the design poured its effort into information and treated the social side as something that would happen on its own.
The trouble is that the social side is precisely the part that decides whether someone stays. A new person's sense of whether they belong here, whether these are their people, whether they can ask a stupid question without judgement, is fragile and forms fast, in the first days and weeks. Leave it to chance and for many people it simply does not form; they remain on the outside, however well the laptop was configured. Microsoft's reform was to stop treating belonging as a happy accident and start treating it as something to be engineered, with the cheapest, most reliable tool available: a real conversation, early, with someone who matters.
This is the difference between informing and designing. Informing works on what the new person knows and assumes the rest will follow. Design shapes the actual experience of arriving, so that belonging is built rather than hoped for.
Turn your instinct for progress into a method
The Behavioural Design Method gives you tools to find the friction in any situation and redesign the path. For your organisation, your team, or your own work.
The science: belonging uncertainty and the fragility of the first weeks
The behavioural principle underneath Microsoft's findings is belonging uncertainty, a concept developed by the psychologists Gregory Walton and Geoffrey Cohen. Their research showed that in any new or evaluative environment, people carry an open, anxious question: do I belong here? In that uncertain state, small cues are read as large evidence. An unanswered email, an awkward first meeting, a sense of being overlooked can be taken not as ordinary friction but as confirmation that one does not fit, and that interpretation shapes everything that follows, from confidence to performance to the decision to stay.[2]
The striking part of Walton and Cohen's work is how small the effective interventions are. In their studies, a brief, well-timed message reframing early difficulty as normal and temporary, rather than as proof of not belonging, produced effects on motivation and achievement that persisted for years. The lever is not large or expensive. It is a timely signal, at the fragile moment, that says you are one of us and the early wobble is just part of arriving. A single early human connection does exactly this: it answers the open question before it hardens into a negative conclusion.
This is why a thirty-minute conversation outperforms a thick onboarding pack. The pack addresses what the person needs to know; the conversation addresses the question they are actually, anxiously holding, which is whether they belong. Microsoft's data is belonging uncertainty made visible at organisational scale: meet the new person early and personally, and you settle the question in the direction of staying.
Through a Behavioural Design lens
It is worth slowing down on what Microsoft actually did, because it is a clean example of how people really settle into a new place, as opposed to how we usually try to induct them.
Start with the person, not the process. Before you can help anyone settle, you have to understand the progress they are really trying to make. A new starter is not, underneath, trying to complete modules; they are trying to find out whether they made the right choice, whether they can do the job, whether these will become their people. Hold that real motive in view and the forces around it come into focus. Two pull a new person towards committing: the discomfort of feeling like an outsider, and the deep reward of being welcomed and feeling they belong. Two hold them back: the anxiety that they do not fit or are not good enough, and the easy option of staying guarded and keeping one foot out of the door. Map those four forces around the real motive and the problem reframes itself. The thing that decides whether they stay is not how much they were told, but whether their belonging was confirmed before their doubt could harden. This way of mapping a behaviour, the deeper motive plus the four forces around it, is what we call the Influence Framework at SUE, and it is where any onboarding should begin: with the human anxiety of arriving, before the information.
Now to the intervention. Microsoft's insight works on the trigger and on belonging: a deliberate early contact, placed at the precise moment the belonging question is most open, answers it before it sets the wrong way. And by making that contact recur, the regular one-to-one, the ongoing buddy relationship, the sense of belonging is reinforced until it holds on its own. The onboarding lesson is the power of a well-timed trigger at the fragile moment, over any volume of information delivered at the wrong one. This approach of identifying the right moment for the right cue is what we call the SWAC Tool at SUE.
This is the move that travels, and it has nothing to do with HR. The first question is never how much information can I give the new person, but what is the anxious question they are holding, and how do I answer it early. Build in the human contact at the fragile moment, and the belonging you were hoping would form on its own often simply does. That, in one line, is the discipline of Behavioural Design here: you do not leave belonging to chance, you design the early human moment that settles it.
What this means in practice
It is tempting to file Microsoft's findings under onboarding programmes and move on, but the same lever sits underneath how anyone settles into anything new, starting close to home.
At the personal level, think of a time you joined something, a job, a team, a group, and either clicked or never quite did. The difference was rarely the quality of the induction; it was usually whether someone made early, genuine contact. Turned outward, the lesson is simple: when you are the one already inside, be that early contact for the next new person. A single deliberate conversation in someone's first days is one of the highest-leverage kind things you can do, because it lands at the exact moment it matters most.
The same move scales up to leading people. When a new hire does not work out, the instinct is to question the hire, the match, the motivation. Far more often the situation never gave them a fair chance to belong. A good leader designs the first week around human connection, a real one-to-one, a named buddy, an early welcome, rather than around forms, because the connection is what determines whether the rest takes. Block the conversation into the calendar the way you would block a client meeting, because it is worth as much.
And it scales again to the organisation. Why does expensive onboarding still leave early attrition high? Rarely because the information was insufficient; far more often because the belonging was left to chance while the process was perfected. Why do some teams retain new people effortlessly and others churn through them? Often the difference is whether early human contact is built in or assumed. In every case the high-leverage question is the same: not how thorough is our information, but how deliberately do we build the early human connection that settles whether someone stays.
What you can design this week
The same move works on welcoming someone yourself, leading a team that takes on new people, or an organisation with a leaky first few months. Three ways to start:
Build one real conversation into the first week. For the next new person near you, deliberately schedule a short, human one-to-one, not an assessment, just contact. Place it early, when the belonging question is most open, and treat it as the most important thing in their induction, because it is.
Be the early contact, do not wait to be asked. Belonging forms before a newcomer feels able to reach out. When you are already inside, make the first move yourself, so they do not have to manufacture the connection from a position of doubt.
Stop perfecting the information and design the welcome. Before you improve another document or module, ask what answers the new person's real question, which is whether they belong. A timely, genuine human moment beats any amount of polished onboarding material.
Microsoft did not make new people stay by giving them more to read. It built in the early human contact that settles whether someone belongs, and let retention follow. That is exactly as useful for welcoming a single colleague, leading a team that hires, or an organisation losing people in their first months, as it is for one of the largest companies in the world.
If you want to think this way about helping people settle and stay, our Behavioural Design training works through exactly this: how to read the real anxiety of a situation, and design the moment that answers it.
Start designing for behaviour, not hoping for it
The SUE Behavioural Design Method teaches you to find the friction, redesign the path, and make the right behaviour the easiest option. In two days live or at your own pace online.
Frequently asked questions
What is belonging uncertainty?
Belonging uncertainty is the concept developed by Gregory Walton and Geoffrey Cohen to describe the open, anxious question that people carry in new or evaluative environments: do I belong here? In that state, small cues are read as large evidence. An unanswered email or an awkward first meeting can be taken as confirmation of not fitting in. Timely, well-designed social contact settles the question before it hardens into a negative conclusion.
What did Microsoft find about onboarding?
When Microsoft analysed what actually predicted whether new employees became productive and stayed, it found the social side mattered most. New hires whose managers had a simple one-to-one with them in the first week showed markedly better outcomes than those who did not. In a buddy programme, new starters who met their buddy more than once in the first ninety days reported becoming productive faster. Early human contact, designed in rather than left to chance, was the lever.
Why does a conversation work better than a thorough onboarding pack?
Because a new hire is not anxiously asking what they need to know; they are asking whether they belong. An onboarding pack answers the first question and ignores the second. A real conversation, early, with someone who matters, answers the second question directly. That is the question whose answer determines whether someone commits or holds back, so it is the place where onboarding effort has the highest return.
1.5 minutes on influence
10,000+ readers · Free · Unsubscribe anytime