Short Answer: Cleaning companies struggle to find cleaners in 2026 because the labor market is dominated by replacement demand. BLS projects 544,800 annual openings across janitors and housekeeping cleaners through 2034, but only about 5,380 a year come from decade-average net job growth. That makes roughly 99% of openings implied replacement needs. At the same time, the foreign-born labor force was 700,000 smaller in June 2026 than a year earlier, and foreign-born workers make up about 33% of janitors and 56% of maids. More applications alone do not fix that math.
Why cleaning companies can't find cleaners in 2026: what the labor data actually shows
BLS projects 544,800 annual openings across janitors and housekeeping cleaners through 2034, and roughly 99% are implied replacement demand rather than net growth. The foreign-born labor force was 700,000 smaller in June 2026 than a year earlier. Here is how I read the numbers—and how I hire against them.
Table of Contents
- ✓1. The short answer: why you can't find cleaners in 2026
- ✓2. The three numbers that explain the shortage
- ✓3. How I built this analysis, and what these numbers do not cover
- ✓4. Finding 1: almost every cleaning job opening is a replacement, not a new job
- ✓5. Finding 2: the foreign-born labor force contracted, and cleaning has a much higher foreign-born share than the overall workforce
- ✓6. Finding 3: cleaning workers are older than the U.S. workforce, which raises replacement pressure
- ✓7. Finding 4: more applications did not mean more hireable applicants
- ✓8. Residential versus commercial: same direction, different math
- ✓9. Where the wage floor actually sits, and who you are really bidding against
- ✓10. What operators misread in these numbers, and the replacement treadmill
- ✓11. How I hire in a shrinking pool
- ✓12. What this data cannot prove
- ✓13. Questions cleaning owners ask about the 2026 labor market
1. The short answer: why you can't find cleaners in 2026
I spent 17 years running a cleaning company.
The lesson I learned is simple: an application is not a hire, and a hire is not a retained employee. More traffic can make a hiring team busier without making the operation better.
The useful question is not "How many people applied?" It is "How many qualified people started, and how many were still there at 90 days?"
2. The three numbers that explain the shortage
Source: BLS projections and Current Population Survey
3. How I built this analysis, and what these numbers do not cover
Where the data comes from
Employment and openings. I use the BLS 2024-2034 projections for SOC 37-2011, janitors and cleaners except maids and housekeeping cleaners, and SOC 37-2012, maids and housekeeping cleaners. The projections, published August 28, 2025, are BLS estimates - not EmployJoy forecasts. (BLS occupational projections).
Foreign-born labor force. I use the monthly CPS series LNU01073395 and the June 2026 Employment Situation Table A-7. The series is monthly, measured in thousands, and not seasonally adjusted. BLS defines foreign born as people living in the United States who were not U.S. citizens at birth. That includes naturalized citizens and noncitizens; the CPS does not separate noncitizens by immigration status. (BLS Table A-7).
Age and workforce composition. I use BLS 2025 annual-average occupation-by-age data, BLS occupation-by-sex, race, and Hispanic or Latino ethnicity data, and EPI analysis of CPS Outgoing Rotation Group microdata. The BLS 2025 annual-average tables are based on 11 months because October data were unavailable during the federal shutdown. I use the BLS demographic table for sex and Hispanic or Latino shares. I use the EPI table only for naturalized-citizen and non-U.S.-citizen shares, which I add to describe the foreign-born share. (BLS age table). (BLS occupation demographic table). (EPI CPS nativity analysis).
Wages. I use May 2025 national OEWS medians for the two occupations and the July 23, 2026 Chicago area release for local major-group wage context. The Chicago comparison uses mean wages for broad occupational groups, not the exact pay offered by one employer. (BLS national wage table).
What it excludes
This article does not measure the applicant funnel inside Fresh Tech Maid or EmployJoy. No proprietary application, no-show, offer-acceptance, or retention dataset was attached to the brief. Where first-party data would materially strengthen a finding, I mark the exact input needed rather than filling the gap with an estimate.
It also does not treat "cleaning" as one occupation.
Commercial janitorial work and residential or hospitality housekeeping overlap, but the job settings, schedules, and workforce mix differ. Keeping SOC 37-2011 and 37-2012 separate prevents one segment from borrowing the other segment's math.
Finally, the CPS introduced updated population controls in January 2026. BLS revised January estimates when February data were published on March 6, 2026. Official values before December 2025 and from January 2026 forward are not strictly comparable across that break.
The chart marks the break directly.
Source: BLS definitions and CPS technical notes
4. Finding 1: almost every cleaning job opening is a replacement, not a new job
Short Answer: BLS projects 351,300 annual openings for janitors and 193,500 for maids through 2034. Yet projected employment rises by only 47,800 janitor jobs and 6,000 maid jobs over the full decade. On a decade-average basis, that implies about 539.4 thousand of 544.8 thousand annual openings - 99.0% - are tied to replacement rather than net growth.
Chart 1. Replacement-related openings dominate both cleaning occupations. The growth component is the ten-year projected employment change divided by ten; the remainder is an implied replacement component.
Source: U.S. Bureau of Labor Statistics, 2024-2034 occupational projections; EmployJoy calculation
The calculation is deliberately simple and reproducible.
For janitors, BLS projects 47,800 more jobs over ten years, or an average growth component of 4,780 jobs per year. Subtract that from 351,300 annual openings and the implied replacement component is 346,520. For maids, the decade adds 6,000 jobs, or about 600 a year, against 193,500 annual openings. The implied replacement component is 192,900.
This does not mean the exact same number of people quit each year.
BLS openings include workers who move to another occupation and workers who exit the labor force. Nor does it mean growth arrives evenly every year. The point is operational: the scale of hiring demand comes overwhelmingly from refilling existing work.
That changes the management question.
If nearly all demand is replacement demand, a job-board budget can only treat the downstream symptom. The upstream work is to reduce preventable exits, improve fit before the offer, and build a pipeline that can refill roles without pulling the owner back into field work.
5. Finding 2: the foreign-born labor force contracted, and cleaning has a much higher foreign-born share than the overall workforce
Short Answer: BLS estimates the foreign-born labor force was 31.872 million in June 2026, down 700,000 from 32.572 million in June 2025. In the same 2025 CPS microdata analysis, EPI reports a foreign-born share of 32.6% for janitors and 55.8% for maids, versus 19.1% for all occupations. The year-over-year decline is real in the published series, but it does not tell us how much came from emigration, labor-force exit, or survey measurement.
Chart 2. The published monthly foreign-born labor-force series fell by 700,000 from June 2025 to June 2026. The January 2026 population-control break is shown because official values on each side are not strictly comparable.
Source: BLS Current Population Survey series LNU01073395 via FRED and BLS Table A-7
Why this matters for cleaning. BLS estimates that foreign-born workers were 19.1% of the U.S. labor force in 2025 and were more likely than native-born workers to work in service occupations. EPI's separate analysis of the same year's CPS microdata reports that 11.5% of janitors were naturalized citizens and 21.1% were noncitizens. For maids, the shares were 17.5% and 38.3%. Adding those categories gives foreign-born shares of 32.6% for janitors and 55.8% for maids, versus 19.1% for all occupations in the same EPI analysis. (BLS annual foreign-born report). (EPI CPS analysis).
That does not mean every cleaning market contracts in the same way.
The BLS annual foreign-born report shows the Midwest had a smaller foreign-born share of its labor force than the West or Northeast. Local pay, transit, immigration patterns, housing costs, and competing employers all change the size of the usable pool. A national contraction is a signal to inspect local funnel data, not a substitute for it.
The wording matters.
The data do not show that 700,000 immigrants left the United States. A person can leave the labor force without leaving the country, and survey participation can change.
The defensible claim is narrower: the published BLS estimate of the foreign-born labor force was 700,000 lower in June 2026 than one year earlier.
6. Finding 3: cleaning workers are older than the U.S. workforce, which raises replacement pressure
Short Answer: The median age was 46.7 for janitors and 46.2 for maids in 2025, compared with 42.1 for all employed people. Workers age 55 and older made up 32.8% of janitors and 31.7% of maids, versus 23.2% of all occupations. That older profile does not prove faster retirement, but it increases the number of roles exposed to retirement and labor-force exit.
Chart 3. Cleaner occupations have fewer workers under 35 and more workers age 55 and older than the overall workforce.
Source: BLS Current Population Survey, 2025 annual averages, Table 11b
The largest janitor age group was 55 to 64. For maids, the 35-to-44 and 45-to-54 groups were nearly tied for largest, with a large 55-to-64 group close behind.
That profile matters because replacement demand includes labor-force exits such as retirement. It also matters for job design: physical strain, route density, equipment, training, and supervisor support can affect whether an experienced worker stays.
The workforce is not only older. It is also segmented.
BLS reports that women made up 86.4% of maids but 36.2% of janitors in the 2025 annual averages. Hispanic or Latino workers made up 49.8% of maids and 35.6% of janitors. A single recruiting message or schedule design will not fit both groups. Residential and commercial operators need separate job offers, not one generic cleaner ad copied across every opening. (BLS occupation demographic table).
7. Finding 4: more applications did not mean more hireable applicants
Without first-party funnel data, I will not claim that Fresh Tech Maid received more applications and fewer qualified candidates.
But the operating distinction is still useful.
An application is only the first stage. The useful supply shrinks as candidates pass through schedule fit, route or site fit, transportation, pay acceptance, response, interview attendance, background or site requirements, start, and 90-day retention.
Source: EmployJoy operating framework
This is the slowdown paradox: total application volume can rise while usable labor supply remains flat.
The scalable metric is not applications per opening. It is qualified applicants per opening, followed by retained hires per source.
That is the point where recruiting data becomes an operating tool rather than a traffic report.
8. Residential versus commercial: same direction, different math
The direction is similar: both segments face replacement-heavy demand, an older workforce, and a labor pool with a high foreign-born share.
The operating math differs.
Commercial owners often lose margin or a contract when shifts go uncovered. Residential owners often lose route density, recurring revenue, and their own time when a cleaner quits.
Commercial hiring must screen site access, shift reliability, certifications, and contract deadlines.
Residential hiring must screen route, transportation, customer-facing behavior, and whether the promised hours are real.
That is why I would not use one turnover benchmark for the entire cleaning industry.
The commonly repeated 200% figure in trade articles does not trace cleanly to a current, national, primary dataset for both segments. The BLS opening data is broader and more defensible, but it still does not replace company-level turnover tracking.
9. Where the wage floor actually sits, and who you are really bidding against
Chart 4. Chicago cleaning wages sit in the middle of several competing hourly occupational groups. Transportation and sales have materially higher group means.
The legal minimum is not the recruiting wage.
Chicago's minimum wage is $17.05 for covered employers as of July 1, 2026. The May 2025 mean for the building and grounds cleaning and maintenance group was $20.87. Food service averaged $18.14, personal care $20.42, healthcare support $20.79, production $24.94, transportation and material moving $25.81, and sales $28.84.
These broad-group means are not direct offers, but they show the alternatives candidates can compare.
For a cleaner, the job is more than the posted hourly rate.
Guaranteed hours, paid travel, route density, stable shifts, supervisor quality, equipment, physical load, tips, benefits, and schedule control all change the value.
A cleaning company that pays near the floor but requires unpaid travel, broken routes, and last-minute schedule changes is not competing at the posted wage.
It is competing at the worker's net value after friction.
I would link this section to a dedicated local wage pillar rather than turn one national labor analysis into fifty city pay guides. The scalable model is one methodology, then local pages or a calculator that refreshes wages, minimums, and competing occupations by market.
10. What operators misread in these numbers, and the replacement treadmill
Chart 5. The replacement treadmill: exits create hiring load; hiring pressure weakens fit; weak fit creates more early exits. Operators can break the cycle at job value, qualification, speed, handoff, and retention.
Source: EmployJoy framework
The first misread is that low growth should make hiring easy.
It does not.
Low growth describes the size of the occupation. Openings describe the flow of workers through it. A large occupation can have little net growth and still force employers to refill hundreds of thousands of positions every year.
The second misread is that more applicants equal more supply.
They do not.
A labor pool is usable only when a person accepts the job conditions, responds, shows up, starts, and stays. Buying more traffic before fixing pay, schedule, route, communication, or supervision can feed the treadmill faster.
The third misread is treating retention as an HR project that begins after recruiting ends.
The data shows the opposite.
When nearly all hiring demand is replacement demand, retention is recruiting capacity. Every preventable exit creates another ad, another screening queue, another interview, another onboarding cycle, and another gap the owner or operations manager must cover.
This does not reverse my earlier position that most cleaning companies have a recruiting problem, not a pure labor shortage.
The national labor pool tightened at the margin. But the volume of openings is still driven mainly by replacement. Fewer candidates can be real while preventable churn remains the larger operating cause.
11. How I hire in a shrinking pool
Source: EmployJoy operating framework
I would start with one metric: qualified applicants per opening.
Then I would connect it to 90-day retention.
A source that produces fifty applicants and no retained hires is not a strong source. A source that produces ten applicants and three retained hires is. That simple connection stops the hiring team from optimizing the top of the funnel while operations pays for the bottom.
And I would keep the process fast without making it careless.
Speed matters because hourly candidates have alternatives. But speed should remove waiting, duplicate questions, and manual scheduling.
It should not remove the human judgment needed to assess job fit, explain the work honestly, and make the final hire.
12. What this data cannot prove
First, the foreign-born labor-force series has a measurement break. Updated population controls apply to January 2026 data, and January was revised when February data were released in March. BLS states that official December 2025 and January 2026 values are not strictly comparable.
The cleanest headline comparison available in the current Employment Situation is June 2025 versus June 2026, but readers should still know that the series changed its population base during the interval.
Second, the CPS cannot tell us how much of the decline came from emigration, changes in legal status, people stopping work or job search, or lower survey response.
The monthly table measures labor-force status among surveyed residents. It does not track each person across the border. The article therefore reports the change in the published estimate and does not turn that estimate into a claim about deportations or departures.
Third, projections are projections.
BLS models expected employment and openings based on economic, demographic, and industry assumptions. The figures are useful for scale and planning, but they are not a promise that exactly 544,800 openings will occur in each future year or that one city will follow the national path.
Fourth, public occupational data cannot prove the slowdown paradox inside Fresh Tech Maid.
That needs a defined first-party dataset. The minimum useful version would include applications, qualified applications, interviews scheduled, interview shows, offers, starts, and 30- or 90-day retention for the same roles and markets across two comparable periods.
Finally, I did not use the 200% cleaning-turnover figure that circulates in trade coverage.
I could not trace it to a current primary national series that covers both commercial janitorial and residential house-cleaning employers with consistent definitions.
That does not mean high turnover is imaginary.
It means the article should not convert a weak benchmark into a precise national rehire estimate. The BLS opening data is the stronger public anchor; company-level turnover is the stronger operating measure.
13. Questions cleaning owners ask about the 2026 labor market
Is there really a cleaner shortage in 2026?
There is a shortage of usable, retained labor for many cleaning employers, but the national data does not show one simple headcount shortage. BLS projects limited job growth and very high recurring openings. The stronger description is a replacement-heavy labor market in which qualified candidates must be found, converted, and retained repeatedly.
Why are there so many openings if cleaning employment is barely growing?
Because openings include replacement demand. Workers move to other occupations, retire, leave the labor force, or leave a specific employer. Across janitors and maids, BLS projects 544,800 annual openings while decade-average net growth is only about 5,380 jobs per year. The occupation can stay nearly the same size while employers keep hiring.
Is immigration causing the cleaner shortage?
Immigration and the foreign-born labor force are part of the supply picture, not the only cause. The foreign-born labor force was 700,000 smaller in June 2026 than a year earlier, and cleaning relies more heavily on foreign-born workers than the average occupation. Pay, schedules, transportation, management, and local competition also shape the usable pool.
What is the average turnover rate in the cleaning industry?
There is no current primary national turnover rate that cleanly covers both commercial janitorial and residential cleaning companies. Trade articles often repeat rates around 200%, but the definitions and source chain are weak. Calculate your own annual turnover by segment, location, manager, and tenure, and pair it with 30- and 90-day retention.
Why do I get many applications but few qualified cleaners?
Applications count interest, not fit. A qualified cleaner must accept the pay, schedule, location, transportation, physical work, and screening requirements, then respond and show up. Track the reason each applicant fails the screen. That turns a vague applicant-quality complaint into a job-offer, source, or process problem you can fix.
Should I raise wages or improve retention first?
Measure both before choosing. When pay is below realistic local alternatives, retention tactics cannot repair the job offer. When pay is competitive but exits cluster under one manager, route, shift, or onboarding stage, fix the operating condition. The right sequence is benchmark job value, diagnose exits, then change the highest-impact constraint.
Does this hit residential or commercial cleaning harder?
Both segments face replacement pressure, but the constraint differs. Maids have a higher foreign-born share and near-zero projected growth. Commercial janitorial employers have a larger employment base and more annual openings, plus shift and site-coverage risk. Use separate funnel and retention dashboards for residential routes and commercial contracts.
How do I find cleaners in a tighter labor market?
Define the real job, measure qualified applicants, contact good candidates quickly, screen job fit before resume polish, and track retained hires by source. More ad spend should come after the job offer and funnel are working. The best source is not the one with the most applications. It is the one with the most 90-day retained hires.
Download the 2026 cleaner labor-market chart pack
Use the charts, definitions, and machine-readable data in this research package to brief your leadership team, compare residential and commercial hiring pressure, and build a local hiring dashboard.
Next step: Run the free cleaner retention scorecard to benchmark the job conditions that may be feeding replacement demand. Get the EmployJoy retention scorecard
About Wells Ye
Wells Ye is the founder of EmployJoy.ai and an experienced cleaning-industry operator. He ran a cleaning company for 17 years and made more than 2,000 cleaning-industry hires. He holds an MBA from the Wharton School, wrote a #1 Amazon book on service-industry hiring, and holds a ForHumanity Independent Certified AI Auditor credential covering AI, algorithmic, and autonomous systems.
Evidence basis and source list
This article analyzes public BLS projections, Current Population Survey age and demographic tables, Occupational Employment and Wage Statistics, and EPI analysis of CPS nativity microdata. The replacement-opening shares are EmployJoy calculations using a stated formula. No Fresh Tech Maid or EmployJoy applicant dataset was used.
BLS occupational projections and worker characteristics, 2024-2034
BLS Janitors and Building Cleaners Occupational Outlook Handbook
BLS/FRED series LNU01073395, foreign-born civilian labor force
BLS Foreign-Born Workers: Labor Force Characteristics - 2025
BLS 2025 occupation-by-sex, race, and Hispanic or Latino ethnicity table
Loading comments...