When Microsoft Outlook and Teams went down in November, a Vancouver office discovered something rare: silence gave way to conversation. An engineer-in-training told reporters that as colleagues stepped away from their screens, the office grew unexpectedly talkative. Online, workers joked to the scrambling Microsoft engineers that they could take their time. Monday would be fine.
The joke landed because an hour without new demands feels like a small victory. But it also exposes a deeper problem in how companies manage knowledge work, according to Sothy Eng, a professor at the University of Hawaiʻi at Mānoa. In a commentary for Fortune, Eng argues that the modern obsession with utilization, leaving no idle hands and no empty calendar blocks, can make teams slower instead of faster.
The utilization paradox
In a standard queueing model, pushing utilization from 80 percent to 90 percent more than doubles expected waiting time, assuming the same task variability and average duration. The math is simple but uncomfortable. When employees are booked to capacity, there is no slack to absorb a question, a surprise, or a fix.
The cost of a conversation that never happens is invisible, but it is steep. A new employee spends three days trying to solve a problem a senior colleague mastered years ago. Two departments trade defensive emails for a week because neither has ten minutes to pick up the phone. Companies pay top dollar to acquire expertise, yet their drive for hyper-efficiency can leave employees without time to share it.
Proximity helps juniors, but it can hurt senior output
Economists Natalia Emanuel, Emma Harrington, and Amanda Pallais recently studied software engineers at a Fortune 500 company and found that sitting near teammates increased feedback on code by 18.3 percent and improved overall code quality. Younger and less-tenured employees benefited the most.
There was a catch. Experienced engineers wrote less code when they sat near teammates. If a company evaluates that senior engineer only by individual output, the time spent helping looks like a productivity drop. Under pressure to hit optimized KPIs, the experienced worker may put on noise-canceling headphones, leaving the junior worker to figure it out alone.

