Five years after the emergency that sent office workers home, the shouting has subsided into something more useful: data. Employers know more about what remote work does to output, collaboration and careers; employees know what it does to their lives. The picture that emerges is neither the utopia nor the collapse each camp predicted, but a set of trade-offs that different jobs strike differently.
The collaboration tax is real but specific
Research on communication networks inside large firms shows remote work making collaboration more siloed: strong ties strengthen, weak ties — the casual connections that carry ideas between teams — thin out. This matters most for innovation and for new joiners building networks. It is also the argument hybrid schedules answer best: co-located days for connection, home days for concentration, rather than either extreme everywhere.
For more coverage of the stories shaping the day, visit our homepage and the latest reporting in Technology.
Careers and the proximity problem
Several studies find fully remote workers promoted more slowly than office peers doing equal work — a proximity bias managers rarely intend but reliably produce. Companies addressing it evaluate on documented outcomes, rotate meeting leadership between room and screen, and audit promotion data by location. The lesson is uncomfortable and useful: fairness in hybrid organisations is an engineering problem, not an attitude.
What workers actually choose
Given the choice, most remote-capable employees pick hybrid — typically two or three home days — and treat flexibility as worth real money in salary negotiations. Fully remote roles attract enormous application volumes, especially from carers and people far from big-city offices. Meanwhile the commute’s collapse has measurably redistributed spending from city centres to suburbs and smaller towns, a slow economic earthquake still unfolding.
Further reading: Stanford research on working from home.
Management is the actual intervention
The strongest predictor of remote success is not the policy but the management: clear goals, written communication norms, deliberate mentoring and meetings designed for mixed presence. Organisations that trained managers for distributed teams report the best outcomes on both output and retention. The office wars, in the end, were never really about offices — they were about whether management could adapt to managing work instead of attendance.
Productivity: better than feared, with conditions
Large studies of remote-capable jobs generally find output per hour holding steady or rising modestly at home, driven by quieter environments and reclaimed commuting time. The gains concentrate in independent, well-defined work; they shrink or reverse for tasks needing dense collaboration, rapid apprenticeship or shared equipment. The honest summary is that “remote work” is not one thing — it is a variable whose effect depends on the work and its management.
Making hybrid work actually work
- Reserve office days for collaboration, home days for depth.
- Write decisions down; hallway knowledge no longer travels by itself.
- Audit promotions and pay by work location once a year.
- Train managers for distributed teams, not just distributed tools.
Why this matters now
Hybrid work has settled into budgets, leases and family logistics, which makes it infrastructure rather than experiment. The decisions being taken now — which roles are location-flexible, how offices are redesigned, how performance is documented — will be expensive to reverse and will shape labour markets for a generation. The evidence base arriving from five years of forced experimentation is therefore unusually valuable: for once, a workplace revolution can be designed from measurements rather than from the preferences of whoever speaks loudest in the boardroom.
Remote work grew up when nobody was watching the headlines: it became a managed practice with known strengths, known costs and known remedies. The organisations thriving with it share one trait — they designed it, rather than merely permitting or banning it.
