How full an office is

The number most often quoted for how empty offices are, that utilisation peaks at 42% on any given day, is an estimate by Strategy Plus, a consulting and design firm. It appears in the introduction of a systematic review, sourced to a 2014 book, and no study in this literature measured it. We could not find a peer-reviewed paper that measures how full an office actually is over time.

The built environment researcher Lisanne Bergefurt built a virtual open-plan office and varied six things at once: screens between workstations, occupancy at 25, 70 or 100%, plants, views out, window-to-wall ratio and colour palette. 214 employees chose between pairs of designs, saying in which office they expected to feel most productive, most concentrated, least stressed, most relaxed and most happy. Every one of the six attributes predicted at least one of those states. A low occupancy rate without screens between desks was among the things that mattered, and so were views out with plenty of daylight and warm wall colours. Plants mattered more than any of them. One interaction ran the other way: women in the sample attached more importance to a 100% occupancy rate for feeling relaxed and happy.

A systematic review by Olivia James screened 10,242 papers and found 31 that compared open-plan and cellular offices directly. Open plan was associated with worse outcomes on health, satisfaction, productivity and social relationships. Noise and distraction, poor privacy, lighting and glare and worse temperature control were the environmental characteristics that recurred. James reads the financial case for open plan as real but offset by costs that do not appear in the property budget.

At one corporate headquarters, 52 employees wearing sociometric badges went from 5,266 minutes of face-to-face interaction over 15 days to 1,492 after the walls came out, a fall of 72%. The management researcher Ethan Bernstein and Stephen Turban measured the same collapse at a second firm, 67 to 71% across 1,830 pairs. Email and instant messaging rose to fill it. Both firms ended with more people in sight of each other and under a third of the face-to-face time.

The engineering researcher Samin Marzban surveyed the result in buildings that had been designed for it. Across 1,403 post-occupancy surveys from 14 open-plan offices in Australia, New Zealand and Hong Kong, ten of them WELL certified and four not, more than 20% of respondents were dissatisfied with the physical environment either way. The five offices with the highest satisfaction on layout, environmental quality, privacy and connection to the outdoors were all certified, and there was no significant difference in health between certified and uncertified buildings.

The occupational health researcher Katarina Wijk asked who loses least when the desks are taken away. 422 employees rated their work environment three months before a move into an activity-based office and nine months after. Satisfaction fell on average, though it rose for some. Managers and people who had already been working in open plan reported the smallest decline; people moving out of private offices reported the largest. Participants without a university degree were more satisfied with both the physical and the psychosocial environment afterwards than those with a degree.

Vacancy, under-occupation and conversion are questions of the property market, and environmental psychology has barely studied what a half-empty floor does to the people still on it. Marzban surveyed 1,403 people and James pooled 31 studies, and neither measured occupancy. Bergefurt varied it with 214 respondents choosing between pictures of a virtual office. Wijk followed 422 people across a move.