People rarely rave about the snack fridge when they describe a job they actually liked; they name a manager who told the truth and a team that covered a bad week. Work that felt like it counted also comes up.
Pew keeps landing on relationships and respect, not the perks that pad a typical job post. That gap should change how you compare employers. If you hire, it should change what you fix first.
So what should you trust when someone says they liked a job?
Coworkers and managers still win the surveys
Pew Research Center's 2023 study of how Americans view their jobs found 67% of workers were extremely or very satisfied with their coworkers. 62% said that about their manager or supervisor.
Those are the warmest numbers in that report. 53% of higher-income workers said the job was fulfilling all or most of the time, compared with 47% of middle-income workers and 40% of lower-income workers. Money changes how the same job feels.
Thing is, a careers page can claim people-first values in a hero image and still hide a manager who never answers. Reviews catch that faster than slogans. Hunt for a named behavior, not a mood.
Respect is uneven, and the gaps are measurable
Pew's 2024 job satisfaction report is blunt about who feels taken seriously. Half of upper-income workers say their supervisor treats them with respect all of the time. 43% of middle-income workers say that. 38% of lower-income workers do.
Degree holders fare better. 50% of workers with at least a bachelor's degree say their supervisor always treats them with respect, and 40% say the same about coworkers.
Satisfaction climbs with age. 67% of workers 65 and older sit in the highly satisfied group, then 56% among ages 50 to 64, 48% from 30 to 49, and 43% under 30. Race follows the same file: 55% of White workers were highly satisfied, compared with 44% of Hispanic workers, 43% of Black workers, and 42% of Asian workers.
The 2023 Pew study also found 41% of Black workers said an employer had treated them unfairly because of race or ethnicity, versus 25% of Asian workers, 20% of Hispanic workers, and 8% of White workers. If you hire, one company-wide score can hide that. If you're applying, read comments from people in roles like yours. Company averages lie in the middle.
Pay is still the complaint people make
Liking a team does not cancel a weak paycheck. A 2024-2025 compilation of U.S. satisfaction stats put high satisfaction with compensation near 30%, the lowest-rated slice of the job in that roundup. That's a third-party summary, not a federal survey. Treat the 30% as a direction, not a law.
Another statistics roundup reports 54% of workers with higher family incomes were extremely or very satisfied with their jobs, versus 41% with lower incomes. Self-employed people scored 62%, against 51% for everyone else. Among employees who left over limited career development, 65% said they couldn't reach their career goals, 63% cited a lack of advancement, and 59% felt the organization did not invest in their skills.
You can love the work and still walk because nobody will promote you, and you can love the work and still walk because nobody will train you, which is the same wound with two names. Growth still shows up when people leave.
Gallup's version of "I liked this job"
Gallup does not run a best-part-of-your-job comment box. It asks whether people have clear expectations, recognition, a chance to develop, and a link to purpose.
Gallup's Q12 research ties those conditions to outcomes. Business units in the top quartile of engagement show 78% less absenteeism and 23% higher profitability than the bottom quartile.
Turnover moves too. High-turnover organizations with stronger engagement show 21% lower turnover. Low-turnover organizations show 51% lower. Best-practice organizations in Gallup's database average about 70% engagement.
Turns out the jobs people praise in exit chats often had those conditions in place, even if nobody used the word engagement.
Flexibility and perks without the brochure language
Schedule control comes up constantly when people describe a job they didn't want to leave. Vendor posts like to stamp one percentage on hybrid work. I'm not going to recycle those.
Look at whether people can actually use the policy. A listing that says flexible and a review that says cameras on, 9 to 6, Slack after dinner are not the same job.
Perks photograph well. They rarely survive as the reason someone stayed. A wellness stipend on top of a decent manager helps. A wellness stipend instead of a decent manager does not.
Culture is the mushiest word in hiring. Job seekers still weigh it when they pick an employer. Connection to the real culture is thinner than the posters. Hunt for who speaks in meetings, whether PTO gets taken, and whether new hires get a real first month. "Great people" on Glassdoor only helps if the next line describes a behavior you can test in an interview.
How to use Glassdoor and job listings without getting spun
Glassdoor is a review site, not a probability sample. People with a story are more likely to write.
Glassdoor's employer guide cites a 2019 U.S. finding that 80% of job seekers say their perception of a company improves after seeing an employer respond to a review. Glassdoor also tells companies to reply to most reviews, not only the easy ones.
You start on the stars, then you catch yourself hunting for whether anyone mentions the hiring manager by name or the actual hours, and that hunt is usually worth more than the average because averages hide one healthy team sitting next to a burnt-out one. Patterns still beat a single rant.
Do this before you hit Apply:
- Open 8 to 10 reviews from the last year in your job family, not the whole company feed.
- Tally how often the direct manager, coworkers, growth, pay, and schedule show up as a plus or a minus.
- Put the posting next to that tally. If they sell unlimited PTO and nobody takes it, believe the reviews.
- Check whether someone from the company answers criticism in public.
- Cross-check the same employer on the job board you're using and on one other source.
A notes app is enough. You don't need a dashboard. Ask interviewers how they handled the last person who needed a schedule change. That answer beats the benefits PDF.
Ask people why they stay, not only why they left
Exit interviews arrive after the badge is already in a drawer. Stay interviews are the earlier conversation: why someone stays, and what would make them look.
AIHR's stay-interview guidance puts the typical session at 30 to 45 minutes. Use a repeated question set. Make it a habit, not a fire drill after two resignations. The manager is often the right person to run it, because the relationship is the topic. If that relationship is the problem, send someone else.
Keep the questions plain. What do you like most about working here right now? When did you last feel proud of a week's work? What would you copy from another team if you could? What would make a recruiter message tempting?
Don't field a survey and then vanish. Gallup built the Q12 so managers sit with the scores and pick a next step with the team. Anonymous pulses help when trust is thin. People remember the follow-up, not the form.
If you want better reviews next year, work on the respect gaps Pew keeps publishing and the career stall that shows up when people walk. To be honest, another espresso machine will not fix a skip-level who never learned names.
A simple comparison before you rewrite the offer or accept one
| What people praise | Why it sticks | Where it breaks | What to verify |
|---|---|---|---|
| Coworkers and manager | Daily experience; Pew 2023 put high satisfaction at 67% and 62% | One bad boss erases perks | Recent reviews in your role |
| Respect | Strong splits by income, education, and race | Slogans hide uneven treatment | Who gets promoted and who speaks |
| Growth | Leavers name goals, advancement, and training | Development that is only a login | Internal jobs, mentors, paid learning |
| Pay and benefits | Hygiene factor; compilations show pay as the weak score | Competitive with no range | The number on the posting |
| Flexibility | Control over time | Fake hybrid, after-hours pings | Whether PTO and hours are real in reviews |
No column wins every search. Younger workers in Pew's data are less satisfied, so a team full of early-career people may need a real promotion path more than another happy hour.
Limits, then one next step
These studies describe U.S. workers in bulk. They do not prove that unlimited PTO saved a specific startup. They do not make Glassdoor a scientific sample.
Vendor percentages about wellness budgets and must-have hybrid rules got dropped when they could not be traced to a primary survey. The boring findings are the reliable ones. People like being treated decently by the humans they see. They like work that goes somewhere. They notice when pay and respect skip their group.
If you're applying this week, score two employers on manager comments, growth mentions, and whether anyone answers reviews. If you run a team, book three stay conversations before you add another perk.