The Office Is Open. Why Did Conversation Move to Email?

An unobstructed office makes colleagues visible. It does not necessarily make them more willing to talk. That distinction matters whenever a workplace brief treats the removal of partitions as evidence of collaboration. Research on organizational behaviour provides an unusually concrete test: what happened to actual interaction when two companies opened their offices?

The architectural question is not whether every open plan is wrong. It is whether visual accessibility, acoustic privacy and permission to interrupt have been confused in the specification of a workplace. A glass enclosure and a desk without partitions are different interventions; neither should inherit an untested promise of better teamwork.

What the researchers measured

Ethan S. Bernstein and Stephen Turban’s 2018 study followed two corporate headquarters before and after conversion to more open layouts. Wearable sensors recorded face-to-face interaction, supplemented by electronic communication records. The first sample included 52 employees; the second, 100. In the first study, measurement covered 15 working days in each phase; the second used eight-week windows.

The quantitative result

Face-to-face interaction time fell by 72% in the first study, while emails sent between participants increased by 56%. In the second, the estimated decline in face-to-face time was 67–71%, depending on the model. These are changes in measured communication, not equivalent percentage changes in productivity. The paper reports a qualitative management assessment of lower productivity in the first company, not an independently quantified productivity loss.

What privacy research adds

Jungsoo Kim and Richard de Dear’s 2013 analysis used 42,764 responses from 303 office buildings. Dissatisfaction with sound privacy reached 59% in high-partition cubicles, 58% in low-partition cubicles and 49% in open layouts with no or limited partitions, compared with 18% in enclosed private offices. Their analysis found that the benefits associated with ease of interaction did not outweigh the penalties associated with noise and privacy in open-plan settings.

This is survey evidence, not a randomized comparison. It nevertheless challenges a simplistic remedy: raising a workstation screen is not the same as providing speech privacy. Equally, a transparent room can be visually exposed while acoustically separated. Designers need to specify the particular boundary being changed.

The limits of the inference

The before-and-after study is not a randomized trial isolating glass, acoustics or visibility. A workplace conversion changes a bundle of conditions. Two organizations cannot establish a universal effect for every occupation, culture or hybrid-working pattern. Sensors record interaction patterns, not whether a conversation resolved a problem; more conversation is not automatically better work. Electronic communication may be appropriate for some tasks.

ArchUp’s interpretation is therefore narrower than a verdict against openness: proximity should be treated as an opportunity for interaction, not proof that useful interaction will occur. A hypothesis worth testing is that control over when one is approachable matters alongside proximity. The cited studies do not establish a single optimal layout or a universal partition height.

A testable architectural response

For an office refurbishment, the proposed intervention is a reversible team neighbourhood with nearby enclosed conversation rooms and quiet work settings, rather than a floor-wide conversion justified by appearance alone. This is an application hypothesis, not a design validated by the two papers. Access rules matter: a retreat that requires managerial permission may not offer practical choice.

Before rollout, define the coordination task that the redesign is meant to improve—for example, resolving cross-team queries without repeated interruption. Establish a baseline, pilot a limited area and compare equivalent work periods. Where feasible, use a comparable team that has not moved, recording occupancy, workload and organizational changes rather than attributing every difference to the plan.

Evaluate perceived privacy, room availability, interruptions and task-specific outcomes together. Use voluntary, aggregated reporting with clear retention limits, not individual productivity surveillance. Measure acoustic performance separately from visibility. Agree in advance what would trigger an adjustment and who can authorize it. These are proposed evaluation procedures; the research does not supply a guaranteed threshold of success.

✦ ArchUp Editorial Insight

The managerial demand to make collaboration visible creates a procurement problem: an open floor can be specified, priced and inspected, whereas a useful conversation cannot be certified at handover. When a brief substitutes unobstructed sightlines for a behavioural outcome, the design team can deliver exactly what was purchased without establishing whether coordination improved. This is an institutional risk suggested by the research, not a finding about the contracts used by the two companies. Its mechanism is a mismatch between the acceptance date and the period over which staff learn to use a workplace. If success is recorded when partitions disappear, later adaptation falls to employees and line managers, while the original brief may escape reassessment. Messaging then becomes a possible compensating infrastructure rather than evidence that physical proximity worked. A different commissioning arrangement would retain responsibility for reviewing work patterns after occupation, reserve a modest adjustment budget and specify privacy and access to collaboration as separate requirements. Such an arrangement would not guarantee better performance; it would make an unsuccessful spatial assumption easier to identify and reverse. The architectural consequence is not a compulsory return to private offices, but a workplace whose boundaries remain adjustable after the opening photograph.

References

• Ethan S. Bernstein and Stephen Turban. “The impact of the ‘open’ workspace on human collaboration.” Philosophical Transactions of the Royal Society B: Biological Sciences, 2018.
• Jungsoo Kim and Richard de Dear. “Workspace satisfaction: The privacy-communication trade-off in open-plan offices.” Journal of Environmental Psychology, 2013.