Turnover in the first 90 days, and what actually drives it

Driver one: they took the job expecting something different. The most common driver, and the one every recruiter hears within a week of a resignation: "I didn't realise it was every second weekend," or "nobody told me how much driving was involved." Usually not dishonesty. A job ad summarises a role in a paragraph; the actual roster and travel only become real once someone is doing it.

It is also the driver most exposed to hiring itself. A vague description before the offer means the candidate agreed to a job nobody actually described.

Largely addressable in hiring.

Ask directly about shift availability, weekend work and travel distance before the offer.

Describe the actual work, personal care tasks and lone work included, not a euphemism.

Record the answer, so there is something to point back to if expectations diverge.

Treat a hesitant answer as information, not a reason to rush the offer.

Driver two: weeks pass between the offer and the first shift. Checks, inductions and paperwork routinely push several weeks between accepting a role and starting it. Nothing keeps the job real in that gap, while other providers keep recruiting the same person. By first shift, some of what made the offer attractive has faded, and a nearer offer may have landed.

It is the same mechanism behind offer-to-first-shift no-shows elsewhere in recruitment, just showing up weeks later as a resignation instead.

Partly addressable.

Start the compliance clock at the interview, not after a separate decision to proceed.

Keep contact going during the gap. A checked-in candidate is still there on day one.

Interview every applicant on arrival rather than in a queue, so the gap starts later.

Measure the gap itself. Most services have never timed it.

Driver three: onboarding is a paperwork week, not a support week. A new starter's first days are usually consumed by inductions, e-learning and compliance sign-off, with little time on what actually matters: who to call if a client seems unsafe, what to do running late, who to debrief with after a hard shift.

This is where a lot of the workforce actually leaves, and it has nothing to do with who was hired. A well-matched candidate can still quit in week two if the first fortnight feels like a manual and a roster and nothing else.

Not a hiring problem.

Pair every new starter with a named, contactable person for the first month.

Separate compliance completion from job readiness. Finishing modules is not the same as being supported.

Check in at day 3, day 14 and day 30, where early disengagement typically shows up.

Fund onboarding as its own line, not a byproduct of recruitment finishing.

Driver four: in-home and community work is genuinely isolating. Residential care puts a new starter alongside colleagues every shift. In-home and community support does not: days can pass between seeing another staff member, working alone in a client's home with no one nearby to ask. For someone used to colleagues around them, that is a bigger adjustment than the role description suggested.

It is real and specific to this part of the sector, and a hiring conversation cannot create or undo it. What it can do is ask about it honestly before the offer, rather than let someone discover it alone in week one.

Partly addressable.

Ask directly whether a candidate has worked alone before and how they feel about it.

Describe the actual pattern: solo visits, driving time, how often they will see a colleague.

Build a real check-in cadence for home and community staff, not the residential default.

Do not oversell the "flexibility" of community work to someone wanting a team environment.