Hiring guideChecked against BLS JOLTS and Oregon BOLI, 15 September 2026

Employer of Choice: What It Actually Takes

Being an employer of choice means people apply to you sooner, accept faster and stay longer than they do at the business down the road paying roughly the same money. For hourly frontline work it comes down to five unglamorous things: a schedule published far enough ahead to plan a life around, pay that clears the local going rate rather than matching it, a supervisor who is not the reason people leave, a hiring process measured in days, and a first 90 days that does not abandon people. Free lunches, swag, values posters and award badges move none of those. Most of it is measurable from outside: in our campaign data two employers advertising the same role on the same platform can sit three times apart on cost per applicant, and the cheaper one usually has the better answer on schedule, pay and response time.

Strip out the award language and there is a real definition

“Employer of choice” is HR-conference language, and in most places it appears it means nothing at all — a line in a careers page header, a slide in an internal deck, a badge bought from an awards body. None of that is a claim anyone can check, which is why the phrase survives.

There is a version worth having, duller than the brochure one. An employer of choice, in a frontline labor market, is a business where:

• More of the people who see your job apply to it than apply to comparable jobs nearby.
• The people you offer jobs to say yes, first time, without a counter-offer round.
• The ones who start are still there at 90 days, where most frontline attrition happens.
• Your own staff put a name forward — a referral is an employee spending their reputation on you.

That is the whole definition. It says nothing about your mission statement, and all four are numbers you can collect this month.

The market already prices how attractive you are

Advertising a job has one great virtue: the labor market answers in numbers, fast. We can see what share of the people who click through finish an application — the candidate conversion rate, which we call the apply rate — and what each one cost. Across 891 managed campaigns the median was $13.88 per applicant. It varies enormously by role.

Apply rate is the share of ad clicks that finish an application — not applicants per hire, and not a hire rate.

What the gap between roles tells you, and what it does not

A caregiver application costs $3.76 and 31% of clicks finish one. A CDL application costs $26.86 and 8% finish. Do not read that sevenfold gap as a verdict on caregiving employers versus trucking employers. Most of it is supply: far more people can start caregiving next week than hold a current Class A and will be away from home. When the pool is small and already employed, every applicant costs more. That is the market, not you.

The gap that is about you

The column to look at is the middle 50%. Technician and mechanic roles are our largest sample at 145 campaign-months, and the middle half of those campaigns landed between $7.40 and $23.58 per applicant. Same role, same platform, roughly a threefold spread, and that is only the middle half. Skilled trades run $7.58 to $21.49. Registered nursing runs $12.76 to $34.84.

Not all of that is the employer. Geography, budget and creative all matter. But part of it is that some employers have a job that is easier to say yes to and can say so in the ad. A posted shift pattern, a real wage range instead of “competitive pay”, and a promise to call back within a day are things the more attractive employer can advertise and the other one cannot.

The five things that actually move it

Roughly in the order an hourly worker weighs them.

1. A schedule published far enough ahead to plan around

For anyone with children, a second job, a class, or a parent to drive to dialysis, an unpredictable schedule is not an inconvenience. It is what makes a job impossible to keep. Someone who finds out on Thursday what they are working on Saturday cannot arrange childcare. They will take a job that pays less and tells them in advance, and they are right to.

Some states make this a legal floor. Oregon requires large employers — 500 or more employees worldwide — in retail, hospitality and food service to give a written schedule at least 14 calendar days in advance of its first shift, to pay extra when the employer changes it afterwards, and not to schedule anyone within 10 hours of their previous shift unless they agree. Several cities have similar ordinances; rules differ by state, industry and headcount.

If you are under every threshold, publishing two weeks ahead voluntarily is the cheapest advantage on this page. It costs manager discipline and nothing else.

2. Pay that clears the local rate, not pay that matches it

Matching the going rate buys nothing. If four employers within fifteen minutes all pay the same, nobody moves and the tiebreaker becomes the drive and the supervisor. Clearing it by a visible margin shows up in the ad: a posted range beats “competitive pay” every time, because candidates read “competitive” as “less than I make now”.

The objection is that this is the most expensive lever. True, which is why the alternative should be priced honestly. Work out what one departure costs in advertising, manager hours, training and lost output — our turnover cost calculator does this — and compare it against a dollar an hour across the crew. In a seat you refill three times a year, the raise usually wins.

3. A supervisor who is not the reason people leave

Frontline workers rarely quit a company. They quit a shift lead who plays favorites with overtime, a DON who does not answer the phone at 5am, a foreman who corrects people in front of the crew. This is the hardest lever here: personal, slow, and the person causing it is usually good at the technical work.

Look at leaver rate by supervisor, not just by site. If one unit, shift or fleet accounts for a disproportionate share of departures, you have a management problem in one place and no amount of advertising fixes it. Check turnover rates by industry first to see whether your overall number is even unusual.

4. A hiring process measured in days

Most employers refuse to believe speed is an employer-of-choice lever. A frontline candidate usually applies to several places in one sitting, on a phone, often while employed. The employer who calls back that afternoon is not competing with the one who calls in nine days. They already hired the person.

Three things cause most of the delay and none cost money to fix: nobody owns the inbox at weekends, interviews are booked by phone tag instead of a link, and offers wait for a weekly meeting. Fixing the second alone takes days out of time to fill — see interview scheduling. The process is the only preview a candidate gets of how you run things — see improving your candidate experience and the glossary entry for candidate experience.

5. A first 90 days that does not abandon people

The cheapest retention you will ever buy is in the first three months, because that is where most frontline departures happen and nothing prevents them except attention. A named person responsible for the new hire. Equipment ready on day one. Check-ins at week one, four and twelve. A written answer to “when is my first raise, and what has to be true for it”.

If you lose people at week three, the problem is not recruitment. It is the gap between the promise made in the interview and the job as run — see first 90 days turnover.

What does not move it

Said plainly, because money goes here:

• Free lunch, coffee, pizza Fridays. Nobody ever took a worse schedule for a sandwich.
• Values posters and culture decks. “We treat each other with respect” on a break room wall, where the schedule drops on Thursday, is read as a joke.
• Branded swag. Pleasant for someone who already likes working for you; nothing to someone deciding whether to apply.
• Annual engagement surveys with no action after. Worse than neutral — asking what is wrong and changing nothing teaches people that telling you is pointless.
• Award badges, and ping pong tables. Nearly all published advice on this subject was written for salaried office employers and does not survive contact with a 5am start.

None of these are bad things. The failure mode is doing the cheap visible ones instead of the expensive invisible ones, then wondering why nothing moved.

What an “employer of choice award” is actually worth

Mostly it is a paid marketing exercise, and it is worth being specific about why.

Many “best places to work” and “employer of choice” programs are run by business journals and awards companies. The model is usually some combination of an entry fee, a table at a gala, a license fee for the logo and an advertising package in the issue announcing the winners. Most have a nomination form rather than a methodology, and the winners closely resemble the entrants.

That is not the whole picture. Legitimate survey-based programs exist and look different: they send a questionnaire to your employees, require a minimum response rate, publish the scoring, and it is possible to enter and fail. Those produce something useful — not the badge, but the data telling you what your own people said anonymously.

Four questions that separate the two

1. Were your employees surveyed, and how many? If the award is decided from a form filled in by your HR director, it measures your HR director.
2. Did you pay to enter, and what does the logo license cost? Paying is not disqualifying — real surveys cost money — but a fee plus no survey plus an ad upsell is a purchase, not an award.
3. Is the methodology published, and can you fail? If every entrant wins, the information content is zero.
4. Would the result change anything you do? If it only goes on the careers page, you bought a graphic.

Be realistic about the audience. A CNA scrolling her phone at 9pm is weighing the wage, the shift and the drive, not a badge from a business journal. Awards work on other employers and your board, not on the candidate.

The levers ranked by cost against likely effect

Cost means money and management attention. Effect is our judgment of how much a lever plausibly moves the four measures above for hourly roles. It is a ranking, not a study.

How to tell whether it is working

Four numbers, all available without new software, all of which move within a quarter if something real changed. What to compare each against is on our hiring metrics benchmarks page.

Measure by role family and by site, not company-wide, or a good unit hides a bad one.

What normal turnover looks like before you panic

Some quitting is just the market working. In the BLS JOLTS release for July 2026, published 1 September 2026, 3.1 million people quit a job in a single month — a quits rate of 1.9% of total nonfarm employment. The industry detail shows how unevenly that sits: accommodation and food services ran 3.5% for the month, retail trade 3.1%, trade, transportation and utilities 2.6%, health care and social assistance 1.9%, construction 1.9%, manufacturing 1.4%. July 2026 figures are preliminary.

A restaurant and a machine shop should not hold themselves to the same number. Judge it against your industry first, then your own sites.

Where branding fits, and the order to do this in

Employer branding is how you describe the job, and it is downstream of the job being worth describing. Photograph a crew that is short-staffed every Saturday and you have advertised a job people leave. That said, most employers under-describe a decent job rather than over-describe a bad one, and that fix is cheap. We keep the tactical side separate: employer branding on a budget covers what to do with little or no money, the glossary entry for employer branding defines the term, and our roundup of five employer branding examples shows what good looks like.

A workable order of operations

1. Weeks 1–2. Pull the four numbers for the last two quarters, split by site and role, and put a real wage range and the shift pattern in every open ad. Do not act on anything else yet.
2. Week 3. One named person owns applicant response, to a one-business-day standard, with a booking link instead of phone tag.
3. Month 2. Publish schedules two weeks ahead at one site and hold the line for a full quarter before judging it.
4. Month 2. Write the 90-day plan: named buddy, day-one equipment, check-ins at weeks 1, 4 and 12.
5. Month 3 on. Deal with the supervisor people are leaving, then price the raise against the turnover it would prevent. Then, and only then, spend on branding or awards.

If you do the first five and nothing improves, the problem is pay or the supervisor — and you now have the evidence to argue for the budget.