How We Reduced Employee Turnover by 34% with Better Scheduling

A janitorial company in the Midwest was burning through cleaners at 280% annual turnover when they called us in. They were paying close to market rate, running a clean operation, and still losing roughly three out of four cleaners every year. Six months later, that number was 185%. Same pay, same management team, same buildings. Three changes to how they ran scheduling.

This is the case study. Twenty-eight sites, 140 employees, real numbers, what worked and what didn't.

The industry context that made this urgent

Commercial cleaning has the worst retention profile of any large service industry in the United States. BSCAI puts annual turnover in commercial cleaning at roughly 200%, with mid-size operators routinely seeing 280 to 350%. ISSA labor cost research shows that replacing a single cleaner runs $1,000 to $5,000 once you factor in recruiting, onboarding, and lost productivity during the gap.

For a company with 140 employees and a 280% turnover rate, that's roughly 392 hires a year. At an average replacement cost of $2,500, the annual replacement bill alone was around $980,000. That number was visible to the owner, but the way it kept showing up was as undertrained crews, missed deadlines, and increasingly unhappy clients. Turnover wasn't a line item. It was the reason every other operational metric was getting worse.

What we thought the problem was

Going in, we assumed the issue was wages. The team was paying $14 an hour in a market where competitors were starting to offer $15.50. We expected exit interviews to confirm that, recommend a wage adjustment, and move on.

That isn't what the data said.

What the exit interviews actually revealed

Over six weeks, we ran structured exit interviews with 31 departing employees. The reasons they gave for leaving fell into a clear pattern.

Rank	Reason	Share	Common quote
1	Unpredictable schedules	42%	"I never know my hours until the weekend before"
2	No input on shift assignments	28%	"I told them I can't work Tuesdays, they keep scheduling me"
3	Favoritism in site assignments	19%	"The same people always get the easy buildings"
4	Poor communication from management	7%	"Nobody tells you anything until you mess up"
5	Pay	4%	"I got an extra dollar somewhere else"

Pay was fifth. Three of the top four reasons were the same problem dressed differently: scheduling that didn't respect the worker's life. People weren't quitting because they couldn't pay rent. They were quitting because they couldn't plan their week.

The three changes we made

Change 1: Schedules published 14 days in advance

The company had been publishing schedules every Friday for the following Monday. We moved them to 14-day advance publication. Cleaners now saw their next two weeks every Monday morning.

This sounds like a small operational tweak. It wasn't. For a workforce that mostly juggles a second job, has childcare to arrange, or relies on rides to work, "what are my hours next Wednesday" is one of the most-asked questions of their week. Answering it on the previous Monday, instead of the previous Friday, removes the single biggest source of friction in the relationship.

Within 90 days, the call-out rate fell from 12% to 4%. That's a 67% reduction in absenteeism, achieved without changing a single hiring or supervision practice.

Change 2: Preference-based assignment with transparency

Each cleaner filled out a short preference card. Preferred sites, days they couldn't work, shift times that fit their life. The scheduler used those preferences as inputs, not afterthoughts. When a preference couldn't be honored, the cleaner got a short explanation rather than silence.

The hard part wasn't building the system. It was the discipline to actually use it. Schedulers, when under pressure, default to the easiest assignment, which is often the one that ignores stated preferences. Holding the line on respecting input was the first month's main management task.

By month four, the share of cleaners reporting "my schedule respects my needs" rose from 31% to 78% on monthly pulse surveys.

Change 3: Documented rotation on undesirable shifts

Weekend cleans, holiday coverage, and the two notoriously difficult sites had been assigned by supervisor's discretion. New hires got them. People who'd recently complained got them. Whether that was actually true didn't matter, because the team perceived it that way.

We built a rotation log that was visible to every cleaner. Weekend shifts cycled through the available pool. Holiday assignments were documented a quarter in advance. The two hardest sites rotated monthly.

The rotation didn't make those shifts pleasant. It made them feel fair. That distinction did more for retention than the schedule changes themselves.

Results after six months

Metric	Before	After	Change
Annual turnover rate	280%	185%	-34%
Call-out rate	12%	4%	-67%
Schedule complaints per week	23	6	-74%
Employee satisfaction (1-10)	4.2	7.1	+69%
Estimated annual savings	$0	$182,000	n/a

The $182,000 figure is conservative. It counts only the avoided replacement costs from the turnover reduction, calculated at $2,500 per cleaner times the difference in attrition. It doesn't count the productivity gain from a more experienced average tenure, the reduced overtime spend on uncovered shifts, or the contracts they kept because client complaints went down.

Why scheduling beat pay as a retention lever

The intuitive read on hourly cleaners is that money is the only motivator. The data on this case, and on industry research more broadly, says otherwise. People in service jobs treat schedule predictability as a form of compensation. An unpredictable $14-an-hour job feels worse than a predictable $14-an-hour job, and they'll quit the unpredictable one even when the predictable one pays the same.

That makes sense once you think about it from the cleaner's side. A schedule they can plan around lets them take a second job, attend their kid's appointments, see family on weekends, and treat the work as part of a manageable life. An unpredictable schedule makes the rest of life harder, even when the hours and pay are the same.

What this actually cost to implement

The total project cost was small. Software for the scheduling and rotation tracking, training time for two supervisors, and roughly 40 hours of management attention spread over the first quarter. The hardest input wasn't money. It was the discipline to publish schedules earlier, even when the natural instinct was to wait for one more confirmation.

Companies that try this and revert tend to revert because the early-publication discipline cracks under pressure. Two weeks in advance feels uncomfortable when you're missing a confirmation. The whole gain depends on holding that line.

When this approach won't move the needle

If your turnover is already under 80% and your call-out rate is below 5%, the gains from this kind of change are marginal. You're already doing most of it. If the underlying issue is wages that are 20% below market, no scheduling change will save you. People will quietly leave for the better-paying option regardless of how kindly you treat them.

This case worked because the company was paying market rate, running a competent operation, and still losing people to a problem they hadn't named yet. If that describes your business, the same three changes are worth running.

The takeaway for your business

You don't have to double wages to keep cleaners. You have to treat scheduling as a retention tool, which is something most operators don't. The three changes above cost almost nothing and reliably move turnover by 30 to 40% in the kind of operation we just described.

"Our cleaners stopped asking when they'd know their schedule. They just knew. That small change transformed our retention numbers." HR Director, the company in this case study

If you want to see what predictable scheduling looks like in practice, our complete guide to multi-site cleaning operations covers the operating model. See how CleanLog handles preference-based scheduling out of the box.