The First Mile of Hiring: Why You Keep Paying to Replace Caregivers You Already Hired

I was talking to an agency owner a few weeks back who felt good about her recruiting, and she had reason to. She'd hired 63 caregivers the year before. Sixty-three. For an agency her size, that's a real number, and she said it like a win. Then I asked how many caregivers were on her roster now. About 70. She'd hired 63 caregivers, and her team was maybe seven people bigger than the year before. The rest had walked. Some in the first month, some after six, a few she barely remembered onboarding. Sixty-three hires, and she was mostly bailing water, working flat out just to keep from sinking. What stuck with me is that she didn't see it that way. In her head, 63 hires meant the recruiting machine was working. What it really meant was that she'd spent a year and a small fortune to land almost exactly where she started.

Three Out of Four. Every Year.

If you've spent any time in homecare, the turnover numbers won't shock you, but they're worth saying out loud. The latest Activated Insights benchmarking report, from the group that used to be Home Care Pulse, put median caregiver turnover at 75% in 2024. That's actually down; it was pushing 80% the year before. Sit with it anyway. Even after the improvement, three of every four caregivers leave in a given year. Now put a price on it. Replacing a single caregiver runs around $2,600, and that's the conservative number. Add the job ads, the screening, the background checks, and the hours your coordinator burns chasing all of it, and plenty of agencies are closer to $5,000 a head. Run that against my friend with the 63 hires. Say 55 of them were backfills for people who left. Fifty-five times $2,600 is north of $143,000, in a single year, spent to stand still. And $2,600 is only the part you can see. The part you can't is the referral she turned down last Tuesday because she had nobody to staff it — the revenue that walked because the caregiver who could've taken that case quit in March and the replacement wasn't on board yet. Turnover doesn't just cost you the rehire. It costs you the work you can't say yes to.

Hiring Has Its Own First Mile

Here's how I've come to think about it. There's a stretch in every agency I'd call the first mile of hiring: the distance between "we need a caregiver" and "we have a caregiver who's trained, working, and not already looking around." Bolt built its whole approach around the first mile of care, the pre-service work that quietly decides everything downstream. Hiring has a first mile too, and it's where the money leaks out. Most agencies run that mile badly, which is exactly why they end up running it so often. It's slow, it's leaky, it's held together with spreadsheets and good intentions. A broken first mile doesn't just cost you a hire. It costs you that same hire again a few months later. So where does it break? Usually in three places.

Start with the application. A good caregiver doesn't apply to your agency. She applies to five agencies in one afternoon, from her phone, on a break between shifts. Whoever answers first usually wins her. If her application lands in a shared inbox somebody checks "when things calm down," she's already accepted another offer, and you'll be paying to go find someone just like her next month. I've watched agencies sit on applications for two and three days and then ask, straight-faced, why their applicant numbers look so rough.

Then there's the stretch between "yes" and "first shift." She said yes. Now there's a W-4, an I-9, a license to verify, a stack of forms that need a real signature, and a coordinator playing phone tag to round it all up. Every day she sits in that limbo is a day she's still taking other agencies' calls. I wrote a whole post on onboarding speed so I won't rehash it, except to say this: the people you lose in that gap are the most expensive ones, because you'd already done the hard part of landing them.

And then the first 90 days, which is the part that quietly eats your math alive. The benchmarking data says 57% of caregiver turnover happens in those first three months. It's rarely the work itself: it's that the new caregiver feels like a stranger. She texts and nobody answers. She's not sure who to even call. She got hired into a black hole, and by week six she's back on Indeed while you're back at the top of the funnel with your wallet open.

What Actually Changed for Her

I'll tell you what shifted for the owner I started with, and the fact that it wasn't dramatic is sort of the whole point. She didn't fix her turnover. Nobody fixes turnover with software. The wages are the wages, the labor market is the labor market, and any vendor who promises you otherwise is selling you something. What she fixed was the bleeding she was doing for no good reason at all. She got every application into one place instead of three inboxes and a spreadsheet, and she switched on texting, so a caregiver who applied at 8 p.m. heard back that night instead of catching a voicemail Thursday afternoon. Answering people while they still cared pulled more of them through than any new job board ever had. The forms started moving through a single workflow with reminders that sent themselves, so the gap between "yes" and "first shift" stopped eating a week at a time. And new hires landed somewhere they could actually reach a person — a portal with real messaging, case openings going out to the whole roster at once, the simple feeling of having joined an agency instead of vanishing into one. None of it is flashy. It's mostly just not dropping people. Those are the two pieces of Bolt that handle exactly this stretch: an applicant tracking system to stop losing candidates at the front, and a caregiver portal to keep a grip on them once they're in. Unglamorous work, making sure the people you fought to hire don't slip out a side door you didn't know was open. Last I heard she was only a few months in. Her turnover hadn't magically dropped by half. But she was answering applicants the same day, fewer new hires were disappearing in the first month, and for once her roster was growing on fewer hires instead of more. The leak was closing.

Back to the 63

That's really the whole thing. She was never a bad operator. She's a good one who'd been losing people in spots she couldn't see, which honestly describes most of the agencies I sit down with. The answer was never to recruit harder; she was already recruiting plenty. It was to quit pouring her best hires into a bucket with a hole in the bottom. If you ran 60-some hires last year and your roster barely moved, your problem probably isn't recruiting. It's your first mile. And unlike the labor market, that part is actually in your hands.