The DoorDash driver can't deliver tomorrow's orders today

Every few months, a new work-life "fix" makes the rounds: no meetings before 10am, no emails after 6pm, unlimited PTO, a four-day week, a Results-Only Work Environment where nobody cares when you log on as long as the work gets done. I like a lot of these ideas. Some of them are genuinely well-supported in the empirical research.

But here’s the problem that I found with these studies: who were these initiatives designed to benefit in the first place?

A results-only culture only works if your job lets you define "results" as something other than "I stood at this register for eight hours." A grocery store clerk cannot scan groceries ahead of schedule. A DoorDash driver cannot deliver tomorrow's orders today. If your paycheck is hourly or piece-rate, "work whenever, just get it done" is not flexibility, it is a policy that does not describe your job at all.

That gap is the subject of a paper I just published in Community, Work & Family, a “Voices” piece as part of their Big Questions in Work-Family special issue. The short version: most of what we know about work-life interventions comes from studying salaried, schedule-autonomous, telework-eligible employees — then we quietly assume it generalizes to everyone else. Specifically, I focused on four structural features of work that are common across roughly half the US workforce and that most existing interventions simply do not account for:

Inability to telework. Remote work jumped from 20% to 71% of the workforce during COVID, but that gain was wildly uneven. Seventy-six percent of lower-income workers held jobs that could not be done remotely, compared to 44% of upper-income workers. There have been few, if any, studies testing whether telework-based interventions even work for employees in jobs that simply cannot be performed remotely (e.g., grocery store clerk, front-line nurses).

Non-salary pay structures. Hourly and piece-rate work make "results-only" thinking nearly incoherent — time spent is the “result” in such job structures, making it impossible to work ahead or work faster. Moreover, they also prevent reduced-load work or unpaid leave as real options, since both require the ability to absorb a pay cut, which is often impossible for those working in entry-level, lower-pay hourly roles.

Customer-driven scheduling. Retail, food service, and hospitality schedules are set by foot traffic, not by the worker (or even by the manager). A "no weekend email" policy is not much comfort to someone whose shift was scheduled, and can be cancelled, the night before, simply due to fluctuations in customer demand.

The rise of alternative work modes. As much as 36% of the US workforce is now in gig or contract work (think DoorDash and Upwork). Most work-life interventions, especially anything involving supervisor support or workplace training and organizational culture development, assume a stable employee-supervisor or employee-organization relationship. A gig worker may have dozens of supervisors, or none.

The biggest concern I have is that these jobs that carry these constraints are disproportionately held by lower-income, non-white, and female workers. This makes it a labor equity concern that seems grossly overlooked.

To be clear, my argument is not that work-life research is wrong. It’s just that we have built and validated our best interventions for the people who perhaps need them least, and then treated everyone else's job as an edge case. I think that gets the priority backwards, and it may be part of why our field sometimes reads as out of touch to the practitioners actually managing these workers.

If you manage or study hourly, gig, or customer-facing workers, I would love to hear what you think I got wrong, or what I missed. This is a short “opinion”-style piece, not a full empirical study, and it was written to hopefully start a discussion.

Zhou, S. (2026). Who are we helping? The impracticality of work-life initiatives for employees facing structural job constraints. Community, Work & Family. https://doi.org/10.1080/13668803.2026.2709339