Most onboarding is run as paperwork and information transfer, measured by completion. But the first 90 days are really a behaviour and belonging window. The behaviours a new joiner adopts early, whether they speak up, ask for help, or run a first hard conversation, tend to stick. Treat onboarding as behaviour change, with rehearsal of the real early moments, not just policy sign-off.
Ask most organisations how onboarding is going and you will get a completion rate. Contracts signed, systems provisioned, mandatory modules ticked, the policy pack acknowledged. It looks like progress because it is measurable, and it is the wrong thing to measure. A new joiner can finish every form in week one and still not know whether it is safe to admit they are lost in week six. Onboarding is run as administration. What it actually decides is behaviour.
The First 90 Days Set the Defaults
The early-tenure period is not a grace window before the real job starts. It is when the new joiner learns how things are done here, by watching, copying and testing. Do people speak up in this meeting or stay quiet. Is asking for help read as initiative or as weakness. When something goes wrong, who says so and how. These are not in the handbook, and they are settled fast. The CIPD's guidance on induction is blunt that first impressions matter: an employee's earliest experience of an organisation has a significant impact on how well they integrate and how satisfied they are. The defaults a person picks up in those weeks are the ones they keep.
Decades of onboarding research point the same way. Talya Bauer's work for the SHRM Foundation, Onboarding New Employees (2010), frames effective onboarding around outcomes that are behavioural and relational, not clerical: role clarity, confidence, social acceptance and knowing the unwritten culture. None of those arrive through a form. They arrive through what the new joiner does, and what happens when they do it.
Read It Through COM-B
If onboarding is behaviour change, it helps to use a behaviour-change model rather than a process map. The COM-B model, set out by Michie, van Stralen and West in Implementation Science in 2011, says any behaviour needs three things present at once: capability, opportunity and motivation. Hold a new joiner against it and the gaps in standard onboarding show immediately.
Capability. Does the person actually know how to do the early-tenure behaviours we want, not in theory but in the moment. Telling someone we have a speak-up culture does not give them the words to disagree with a senior colleague. Our behaviour change training treats capability as something you build by doing, not by being told it exists.
Opportunity. Does the environment let the behaviour happen. A new joiner who asks a question and gets a curt reply learns, correctly, that questions are unwelcome here. Opportunity is set by managers and teammates in the small responses of the first month, which is why onboarding cannot be outsourced to a portal.
Motivation. Does the person believe the behaviour is worth the risk. Speaking up has a cost, and a new joiner with no standing weighs it carefully. Motivation comes from seeing the behaviour modelled and rewarded, not from a values poster.
Why Information Transfer Will Not Do It
The reason a content-led induction changes so little is the reason most corporate learning changes little. Being told something, even being tested on it, is not the same as being able to do it under pressure. Roediger and Karpicke showed in Psychological Science in 2006 that actively retrieving and practising material lifts long-term retention by roughly 50% compared with simply re-reading it. A welcome deck and a policy quiz are the re-reading kind of learning. They feel productive in week one and have faded by the time the moment that matters arrives.
And the moments that matter in early tenure are behavioural: the first time you flag a slipping deadline, the first disagreement with someone senior, the first time you do not understand and must decide whether to say so. The work that genuinely shifts psychological safety is built on the construct Amy Edmondson defined in 1999, the shared belief that the team is safe for interpersonal risk-taking. You do not install that belief with a slide. A new joiner forms it from what happens when they, or someone next to them, takes the risk.
What Onboarding for Behaviour Looks Like
Because real behaviour change happens through lived experience that makes the memory stick, we build onboarding around rehearsal of the real early-tenure moments, not policy sign-off. New joiners practise the actual situations, raising a concern, asking for help without apologising for it, running a first difficult conversation, with people who push back like real colleagues do, and they get honest feedback on what they did. The same moment is run before and after, so the change is observed rather than self-reported. Our employee workshops and broader training and development are designed for this, so a cohort leaves having behaved differently in the room, not just heard how they should.
This matters for managers too, because they set the opportunity half of COM-B every day. A line manager who runs the first one-to-one well, and answers the first question in a way that invites the next one, is doing more for onboarding than any portal. That capability is rehearsable like any other.
Onboarding is not a checklist to complete. It is the window where a new joiner's behaviours are set. Measure what they do by day 90, not how many forms they signed in week one.
So the honest test of an onboarding programme is not its completion rate. It is whether, ninety days in, the new joiner speaks up, asks for help and handles the hard moment, because that is what the first ninety days quietly taught them to do. If your onboarding is measured by sign-off, you are measuring the paperwork and missing the point.
Book a free 30-minute diagnostic call → or read about our research-backed approach.
Sources: Michie, van Stralen & West, “The behaviour change wheel: a new method for characterising and designing behaviour change interventions”, Implementation Science, 2011 (COM-B: capability, opportunity, motivation). Roediger & Karpicke, Psychological Science, 2006 (the testing effect). Edmondson, Administrative Science Quarterly, 1999 (psychological safety). Bauer, Onboarding New Employees: Maximizing Success, SHRM Foundation, 2010. CIPD, Induction factsheet. Get in touch today. We are Sidestream.