The first-90-days problem: why new joiners leave

Most retention conversations focus on people who have been around for years: the senior engineer who got a counter-offer, the manager who burned out after a long stretch. But for many Indian organisations, one of the highest-attrition windows is not year three or year five. It is the first 90 days. New joiner attrition in the first 90 days is the churn that tends to hurt most, because you have spent on hiring and onboarding and recovered almost nothing in return.

This is also the window where intervention is cheapest and most effective. A new joiner who quits in month two rarely does so over one dramatic event. They drift: an unclear role, a manager who is too busy to check in, a sense that this place is not what was promised. None of that shows up in an exit interview until it is too late. The argument of this piece is simple. The first 90 days are fragile by design, the cost of getting them wrong compounds, and a structured rhythm of wellbeing check-ins is one of the few levers that can reliably move the number.

A new joiner arrives with almost no context and almost no relationships. They do not yet know who to ask, what is normal, or whether the gap between the interview pitch and the daily reality is a red flag or just first-week noise. Every organisation has unwritten rules, and for the first few weeks the new person is guessing at all of them. That uncertainty is cognitively expensive and quietly stressful, even for a strong hire.

Indian workforce realities can sharpen this. Many joiners are serving notice elsewhere or have just finished a long, draining notice period. Shift workers and frontline staff often onboard with thin manager contact and little structured support. And because counter-offers and parallel offers are common, a wobble in the first month is not just a morale dip: it can be an open door for the previous employer or a competing one to walk back in.

The result is a window where small frictions have outsized consequences. A delayed laptop, a manager who cancels three one-on-ones in a row, an unclear first project, confusion over PF transfer or ESIC enrolment: individually minor, collectively they can signal that the new joiner does not matter much. That signal, often more than salary alone, is what drives many early exits.

When a new joiner leaves in the first quarter, you do not just lose the salary you paid. You lose the cost of sourcing and interviewing, any recruiter or agency fee, the onboarding hours from IT, HR and the hiring manager, and the productivity of the teammates who stopped their own work to train someone who is now gone. Then you pay most of it again to refill the role, while the team absorbs the gap in the meantime.

Commonly cited industry estimates put the cost of replacing an employee at roughly one-half to two times their annual salary, depending on seniority and role. For a knowledge role in an Indian metro, that can run from a few lakh rupees to well over ten. The more specialised the hire, the higher the multiple tends to be, because the role sits empty longer and the ramp is steeper. Treat these figures as directional rather than exact: the right number for your organisation is the one you calculate from your own costs.

There is a second-order cost too. Early attrition can be contagious. When a team watches two or three new joiners leave in quick succession, the people who stayed may start to wonder what they are missing, and your employer brand on platforms where candidates compare notes can take a quiet hit that is hard to undo.

Almost every employer believes it onboards well. The gap is rarely intent. It is rhythm. A welcome lunch and a Day 1 deck are front-loaded, then the new joiner is left alone precisely when the early doubts set in. A structured cadence fixes the timing problem by putting deliberate touchpoints at the moments that matter: 15, 30, 60 and 90 days.

Each check-in does a different job. The 15-day check is logistical and emotional: is your setup working, do you understand your first project, does this feel like what you signed up for. The 30-day check tests belonging and clarity of expectations. The 60-day check surfaces whether the role and the manager relationship are actually working. The 90-day check is the honest two-way review before the relationship hardens into habit. Spacing them this way helps catch drift early, while it is still a conversation and not a resignation.

The substance matters as much as the schedule. These check-ins should ask about workload, support and psychological safety, not just task progress. Amy Edmondson's research on psychological safety, and Google's Project Aristotle, which identified psychological safety as the most important of the five dynamics that set its most effective teams apart, both point the same way: people tend to perform and stay when they feel safe to speak up, admit confusion and ask for help. A new joiner who cannot say I am struggling in week three may already be a flight risk by week eight.

A check-in only surfaces honest answers if people trust where the answers go. If a new joiner suspects that I am overwhelmed will land in their probation file, they will say everything is great until the day they resign. The mechanics of trust are not optional decoration here. They are the thing that makes the data real.

Under India's Digital Personal Data Protection Act 2023, that trust also has a legal shape: consent, purpose limitation, and clear data principal rights. Tell new joiners what is collected, why, who can see it, and how it is used. Aggregate and anonymise wellbeing signals so managers see team-level patterns, not individual transcripts. Done well, this is not just compliance. It is the precondition for people telling you the truth early enough to act.

The first 90 days are not a soft, cultural footnote to onboarding. For many employers they are the period where a large share of avoidable attrition happens, where each loss can cost a multiple of salary, and where a churned hire quietly raises the odds of the next one. Reducing new joiner attrition in the first 90 days is not about a slicker welcome kit. It is about building a deliberate rhythm of check-ins, asking real questions about workload and safety, and handling the answers in a way people can trust.

Tools like Aha exist to run that rhythm and keep the data anonymised and aggregated, but the principle stands with or without any platform. Decide on your 15, 30, 60 and 90-day touchpoints, decide who owns each one, and decide where the honest answers go. The organisations that keep their new joiners are rarely the ones with the biggest budgets. They are often the ones that noticed the drift in week three and did something about it before week eight.