Hiring for an hourly role rarely feels like a one-time job. You write the ad, run the interviews, train someone up, and often enough, within a year, you’re doing the whole thing again for the same position.

The Shift Worker Longevity Index puts numbers to that. It draws on anonymized, aggregated Deputy data from US businesses over the twelve months from September 2025 to August 2026, covering 38 sub-industries and 32 states.

For each, we measure annual staff turnover (employees who left during the year, compared with the average number employed) and turn it into an estimated expected tenure, a rough sense of how long a role tends to last. Alongside that, we look at how turnover differs by generation, state and industry, as well as the scheduling and shift patterns linked to turnover, and compare Deputy's figures with national separations data from the Bureau of Labor Statistics.

Across US businesses on Deputy, the average hourly job lasts an estimated 1 year and 3 months (15.4 months). However, some industries hold onto people for years while others refill the same roles more than once a year. Dive into the results below. 

Key takeaways 

• Shift workers turn over fast. Across US businesses on Deputy, annual staff turnover runs at 78.2%, an estimated 1 year and 3 months before someone moves on.

• The distance between industries is wide. Hospitals hold onto people the longest of any sub-industry, with turnover of just 26.1% and an estimated tenure of 3 years and 10 months (46.0 months). 

• Fast Food and Cashier Restaurants sit at the far end at 136.4%, more than five times the hospital rate, and high enough that plenty of roles are filled more than once in a single year.

• Healthcare keeps its people even when the days are hard. It has the lowest turnover of any industry at 56.3%, yet also the highest share of shifts that end stressed or frustrated, at 7.7% against 5.6% across all employees.

• More notice tends to go with longer stays. The third of sub-industries that publish schedules furthest ahead have pooled turnover of 61.0%, against 81.8% for the third that give the least warning. 

• Generations split here too. Outside the small sample of Gen Alpha, Gen Z turns over at 93.0%, almost two and a half times the Baby Boomer rate of 40.2%, and moves on fastest in every industry measured.

• Location makes a difference. Maryland (64.1%) and Massachusetts (65.4%) keep staff longest, while Florida has the highest turnover of any state with a firm figure at 108.2%. Louisiana, Missouri and Oklahoma stand out for pairing some of the most stressed shifts with some of the highest turnover.
  
  

Shift worker turnover by sub-industry

Build the schedule on Monday, and it rarely survives the month intact. Someone picks up a second job, someone goes back to school, someone just doesn’t come back after a rough weekend, and the role opens up again. How often that happens depends a lot on the kind of work.

Hospitals sit clearly out front. Their annual turnover is 26.1%, an estimated 3 years and 10 months (46.0 months) in a job. Animal Health (41.1%) and Government (42.1%) come next; however, there’s a 15-percentage-point gap between first and second place.

Healthcare runs right through the top of the table. Five of the ten sub-industries with the lowest turnover sit in healthcare: Hospitals, Animal Health, Doctors office / Medical Clinic, Healthcare Other and Outpatient Care Centers.

At the other end, Fast Food and Cashier Restaurants turn over fastest of all at 136.4%, an estimated 8.8 months in a role, just ahead of Arts, Entertainment & Recreation at 122.9%. That is more than five times the rate inside Hospitals. Any figure above 100% means that, on average, each role was filled more than once across the year, and six sub-industries cross that line, including Fast Food / Cashier Restaurants, Arts, Entertainment & Recreation, Call Centers and Cleaning Services. 

Hospitality itself lands at both ends. Bars come 6th of 38 at 53.9%, while Fast Food / Cashier Restaurants finishes last. 

Shift worker turnover by industry 

Step back from the sub-industries to the five broad industries, and healthcare stands apart again. Its annual turnover is 56.3%, an estimated 1 year and 9 months in a job, the lowest of the five, and well under the average of 78.2% across all US employees. Retail comes next at 70.8%.

Where the industries really part ways is in how much warning workers get. Healthcare publishes its schedules 10.9 days ahead on average, against 7.4 days across all employees and just 5.7 days in hospitality. More notice goes with longer stays, though the two move together in the data rather than one pushing the other.

Healthcare’s shifts are also longer and harder than most. The average shift runs 7.8 hours, against 4.4 in hospitality, and 7.7% of rated shifts end stressed or frustrated, the highest share of any industry. So healthcare holds onto its people even though the work is among the longest and most demanding on the list.

Shift worker turnover by generation

Age is recorded for 40.0% of US employees, and among those workers the generations fall into a clear order: the younger the group, the faster it moves on.

Leaving out Gen Alpha (indicative only: too few workers), Gen Z turns over fastest, at 93.0% a year, an estimated 1 year and 1 month in a job. From there, every older generation stays longer: Millennials at 61.1%, Gen X at 47.9%, and Baby Boomers at 40.2%, an estimated 2 years and 6 months (29.8 months). That leaves Gen Z turning over at almost two and a half times the Baby Boomer rate.

The order holds wherever you look. Gen Z has the highest turnover in every industry, from 67.9% in healthcare up to 125.5% in the “Other” group, which is high enough, at the top end, that those roles are filled more than once in a year. It’s a pattern worth watching, because Gen Z is the largest generation among Deputy employees with a recorded age, at 56.1%. 

Shift worker turnover by state

Maryland holds staff longest, at 64.1% annual turnover, an estimated 1 year and 7 months in a job. Massachusetts sits just behind at 65.4%, with Alabama close on their heels at 65.9%.

It isn’t simply that those states lean toward stickier industries, either. Both Maryland and Massachusetts also come in well below what their mix of industries would predict, each turning over at 0.79 times the expected rate.

At the other end, Florida has the highest turnover of any state with a firm figure, at 108.2%, ahead of Missouri at 101.3%, both running above what their industry mix predicts. Georgia’s 166.1% is higher still, but it’s indicative only, driven by a single sub-industry, so it sits outside the comparison.

Where turnover and stress line up, it’s across the South and Midwest. Louisiana (11.9%), Missouri (10.6%) and Oklahoma (9.2%) record the most stressed or frustrated shifts of any states, and all three also rank among the ten with the highest turnover.

The calmer, steadier end skews the other way. Massachusetts pairs the second-lowest turnover with the fourth-least stressed shifts (3.3%), while Idaho (2.4%) and Colorado (2.5%) have the least stressed shifts of all.

What’s linked to turnover? 

Shift stress 

Deputy’s Shift Pulse asks workers how a shift went, and the first thing that stands out is what doesn’t connect: how a shift feels and how long people stay don’t line up. Across the ranked sub-industries, the share of shifts that end stressed or frustrated doesn’t move with turnover.

Healthcare makes the point plainly. It carries the most stress of any industry, with 7.7% of rated shifts ending stressed or frustrated, against an average of 5.6% across all employees, and still keeps staff longest. Four of the ten sub-industries with the most stressed shifts are in healthcare.

Among sub-industries with a firm figure, Critical & Emergency Services has the most stressed shifts at 10.8%, alongside turnover of 91.1%.

The same healthcare pattern shows up at the sub-industry level, though these figures are indicative only, as they rest on fewer ratings. Animal Health's share of stressed or frustrated shifts is 12.2%, alongside the 2nd-lowest turnover of any sub-industry. Hospitals, with the lowest turnover of all, sit at 6.6%, inside the top ten for stress. At the other end, the calmest shifts are in Gyms, where just 1.1% end stressed or frustrated, ahead of Dentists at 2.5%.

Schedule notice 

Across the ranked sub-industries, more notice tends to go with lower turnover. The twelve that publish schedules furthest ahead (8.4 to 16.9 days) have a pooled turnover of 61.0%. The twelve with the least notice, 3.9 to 6.4 days, sit at 81.8%.

The pattern is largely healthcare against everyone else. Six of the ten sub-industries that give the most notice are in healthcare, which already holds onto staff well on other measures. So this is a link between notice and staying, not evidence that one causes the other. Hospitals show the connection at its clearest. Their schedules go out 16.9 days before the shift, the most of any sub-industry, alongside the lowest turnover of all.

It doesn't hold cleanly everywhere. Delivery & Postal Services (3.9 days) and Construction (4.2 days) give the least notice of anyone, yet rank 8th- and 10th-lowest for turnover. So short notice, on its own, doesn't decide who stays. Giving more notice is still good practice, and in several US cities it's a scheduling compliance requirement.

Shift length 

Across all US employees, the average shift runs 6.2 hours. Healthcare’s are the longest at 7.8 hours and hospitality’s the shortest at 4.4, which is a reflection of how differently the work is organized more than anything else.

The length of a shift tells you little about who stays. The third of sub-industries with the longest shifts (8.0 to 9.9 hours) have pooled turnover of 67.2%, and the third with the shortest (3.7 to 7.0 hours) sit at 80.8%, but across all 38 sub-industries the link is weak, so this isn't evidence that longer shifts keep people.

When do shift workers leave? 

Leaving isn’t spread evenly across the year. Nationally, September is the busiest month for it: 8.9% of US employees left in September, almost twice the 5.0% who left in February, the quietest month.

The end of summer is the heart of it. Three of the five industries peak in September, with healthcare at 6.7%, retail at 7.8%, and the catch-all “Other” group at 13.4%. Hospitality runs a month ahead: 10.9% of its employees left in August, more than twice its February rate of 5.1%, and the widest month-to-month swing of any industry apart from the “Other” group. This suggests many hospitality workers are leaving summer jobs before the school year starts.

Winter is the quiet stretch. February is the calmest month nationally, and every industry bottoms out somewhere between November and February: November for retail and services, December for healthcare, and February for hospitality and the “Other” group.

Services peaks earlier than the rest. Its highest month is June (9.2%) and its lowest November (4.8%). Looking back at January to August 2025, March was the month with the most turnover (8.5%), so the summer peak may not be a fixed habit. 

Star Levandowski, SVP of Operations & People, at Deputy comments: 

“What stands out in this data is the range. Across hourly work, the average job lasts a little over a year, but that average hides two very different worlds. In healthcare, people stay for years, even though the shifts are long and among the most stressful we measured. In fast food and a lot of frontline service work, the same role can turn over more than once in twelve months. 

“Age and timing matter too: Gen Z leaves soonest in every industry, and leaving has a clear season, peaking at the end of summer, when many students head back to school. While it may seem obvious, knowing this pattern is important, as it can help businesses get ahead instead of reacting to staff turnover.

“If I had to give employers one place to start to reduce turnover, it’s creating certainty. Give people more warning about when they’re working, and change it less once it’s posted, as a schedule someone can rely on is worth more than most perks. Spread the late nights and overnight shifts around rather than leaning on the same few people every week, to give staff back some personal time with friends and family. I’d also recommend investing in a new hire’s first month, especially where your team skews younger, to ensure they feel capable and confident in the role. Finally, keep an ear on how shifts actually feel. For example, a quick end-of-shift check shows you where people are wearing down long before they decide to leave.” 

About the data 

The Shift Worker Longevity Index shows how long hourly shift workers stay in their jobs across US industries. It ranks sub-industries and states by annual staff turnover and turns each rate into an estimated expected tenure: how long a worker would stay on average at that rate. The national and industry figures use every US business on Deputy, wherever it's based. The state ranking uses only the 32 states with enough data for a reliable figure: a state comes from the workplace's address, which just under half of employees have, and needs at least 800 employees on an average day and 150 businesses.

It's built from anonymized, aggregated data from US businesses that use Deputy, covering hundreds of thousands of hourly workers over the twelve months from September 2025 to August 2026. A full year evens out the seasons, such as the rise in workers leaving at the end of summer.

For each sub-industry, industry and state, we count the employees who left during the year and divide by the average number employed on a day. We also look at turnover by generation, how far ahead schedules are published, how shifts felt at the end of the day, how leaving changes through the year, and how Deputy's figures compare with the national rate from the Bureau of Labor Statistics. The sections below explain each step in more detail.

What we measured

• Annual staff turnover: employees who left during the year as a share of the average number employed on a day, with customer offboarding (businesses leaving Deputy) removed.

• Estimated expected tenure: twelve divided by annual turnover, in months. It's an estimate of how long a worker would stay at that turnover rate, not a measured length of service.

• Turnover by generation: annual turnover of Gen Z, Millennials, Gen X and Baby Boomers, for workers with a recorded age. 

• Schedule notice: the average number of days between a schedule's final publication and the start of the shift, weighted by the number of schedules.

• Schedules deleted: schedules deleted for every hundred published.

• Average shift length: hours worked per timesheet.

• Stressed or frustrated shifts: the share of Shift Pulse ratings that were "stressed" or "frustrated". Shift Pulse is Deputy's optional end-of-shift check-in, where workers rate how their shift felt.

• Share leaving each month: employees who left in a month as a share of the average number employed that month.

Which industries and states we included

Deputy groups businesses into five industries and 47 sub-industries. A sub-industry or state is ranked only if it has at least 800 employees on an average day and 150 businesses, so that a handful of workplaces can't decide its result: 38 sub-industries (covering more than nine in ten employees) and 32 states qualify. Industry figures include every sub-industry, whatever its size. States come from the workplace's address, recorded for just under half of employees, and generation figures from workers with a recorded age (about two in five employees).

How the figures are calculated

Annual staff turnover compares the number of employees who left during the year, for any reason, with the average number on a business's books on a typical day. Estimated expected tenure turns that rate into time: it's how many months the average worker would stay if turnover held at that level.

When a business stops using Deputy, all of its employees show up as leavers in the same month, though it's the business that has left Deputy, not necessarily its workers. To keep these out, any month in which a sub-industry's share of leavers is more than three times its usual month is replaced with that usual month.

Sub-industries and states are ranked from the lowest turnover to the highest. States are compared with each other rather than with the national figure, because employees with a workplace address on record leave more often than those without, and each state is also compared with the turnover its mix of industries would predict.

Scheduling and end-of-shift ratings are compared with turnover across the ranked sub-industries; these comparisons show links, not cause and effect. The year-on-year comparison covers January to August only, as Deputy's data start in January 2025.

How we tested and validated the results

We checked that the results hold up in a couple of ways. Changing the offboarding cut-off from two and a half to five times the usual month keeps national turnover between 77.2% and 78.5%, with Hospitals first and Fast Food / Cashier Restaurants at or near the bottom every time. We also re-ranked with each month left out in turn: neighboring positions can swap, so small gaps in the ranking shouldn't be over-read. States or Shift Pulse figures based on too few workers are marked indicative only. Deputy's month-to-month pattern also tracks the national rate from the Bureau of Labor Statistics.

Where the data comes from

• Deputy: anonymized, aggregated data from US businesses using Deputy, September 2025 to August 2026 (January to August 2025 for the year-on-year comparison).

• US Bureau of Labor Statistics: Job Openings and Labor Turnover Survey (JOLTS), total separations, not seasonally adjusted, September 2025 to August 2026.

Limitations

• Deputy businesses only: the figures cover businesses that use Deputy, not the whole US workforce, and a move between two Deputy businesses counts as leaving.

• No reason for leaving: turnover includes resignations, dismissals and other exits.

• Expected tenure is an estimate: it comes from the turnover rate, not from measured lengths of service.

• Partial fields: age is recorded for about two in five employees and a workplace state for just under half, so generation and state figures cover those employees only.

• The offboarding rule is ours: it can't tell a business leaving Deputy from a genuine wave of exits in the same month, so a few real exits may be removed.

• One year for most seasons: the September peak and the November and December lows rest on one year; only January to August can be compared with a second year.

Disclaimer

This study is for general information. It's based on anonymized, aggregated Deputy data and shows links between scheduling practices and turnover, not cause and effect. Figures describe US businesses using Deputy and may differ from official statistics.