As AI makes shallow work cheap and notifications make focus scarce, the ability to concentrate has become a real competitive edge — and the antidote to a workforce burning out on its own busyness.
The average knowledge worker is now interrupted once every two minutes. Across a day, that adds up to roughly 275 pings from meetings, emails, and chats, according to Microsoft’s 2025 Work Trend Index. Set that against another well-replicated figure: the 23 minutes it takes, on average, to fully refocus after a single interruption. The arithmetic is brutal. The interruptions arrive faster than the recovery, which means that for many people, deep concentration has become structurally impossible.
This is the notification economy, and it is quietly reshaping what good work looks like. It is also doing two things at once that we tend to treat as separate. It is making focused, original work rarer, and it is burning people out. The argument here is that those are the same problem, and that the companies and individuals who solve it gain an edge on both fronts.
The economics of attention — and why focus got scarce
Attention is the product that the modern internet was built to sell, and the workplace inherited the machinery. Gloria Mark, who has studied attention at UC Irvine for two decades, found that the average time a person spends on a single screen before switching has collapsed from two and a half minutes in 2004 to about 47 seconds today, a figure other researchers have replicated within a few seconds. The basic unit of attention in knowledge work is now under a minute.
The damage is not only the time the interruption steals. Every switch leaves what psychologists call attention residue, a smear of the previous task that clings to the next one and degrades it. Worse, the habit becomes self-sustaining. Mark’s work found that when external interruptions are removed, people begin interrupting themselves at almost the same rate, as if a brain trained to expect a ping learns to generate its own.
The result is a workday that feels full and produces little. Asana’s research has found that knowledge workers spend around 60 percent of their time on the coordination of work, the messaging, searching, and status-chasing, rather than the skilled work they were hired to do. Busy and productive have quietly come apart.
Why focus became a competitive advantage
For most of the knowledge-work era, the ability to concentrate was a nice-to-have, a personal productivity virtue. In 2026, it has become something sharper. The reason is artificial intelligence.
AI is extraordinarily good at precisely the shallow, routine cognitive work that used to fill a professional’s day. It drafts the email, summarises the report, formats the deck, writes the first pass of the code. As that work gets automated and AI fluency becomes a baseline skill rather than a differentiator (the World Economic Forum already finds workers with AI competencies earning a clear wage premium over peers without them), the premium shifts to the things AI cannot do for you. Judgment. Synthesis. The taste to see what actually matters, and the skill to frame the right question in the first place. All of it requires sustained, undistracted thought. All of it is the output of deep work.
Deep work is the ability to focus without distraction on a cognitively demanding task.
— Cal Newport, Deep Work
This is the quiet inversion at the heart of the moment. The cheaper machines make shallow work, the more valuable deep work becomes, because depth is now the scarce ingredient that separates commodity output from the kind only a concentrated human mind produces. When everyone holds the same tools, the differentiator is the thinking, and thinking does not survive a ping every two minutes.
AI made shallow work cheap. That is exactly what made deep work valuable.
The other half of the bill — burnout
The competitive case is only half the story. The same notification economy that erodes deep work is also exhausting the people inside it, and that is the more urgent cost. Microsoft’s data describes an “infinite workday” with no clean edges: chats sent outside working hours up 15 percent year on year, meetings after 8 pm up 16 percent, nearly a third of workers back in their inboxes by 10 pm. Almost half of employees and more than half of leaders say their work feels chaotic and fragmented, and a striking 80 percent of the global workforce says they lack the time or energy to do their jobs.
This is not a soft concern. The World Health Organization classifies burnout as an occupational phenomenon, and one in three workers now describes the pace of the past five years as unsustainable. The bill lands on the business as much as the person, in disengagement, mistakes, lost innovation, and the slow departure of the most capable people, who always have the most options. Fragmented attention makes a workforce less productive and more depleted at the same time. That is why fixing focus is not a trade-off between performance and well-being. It repairs both at once.
The pattern takes a particular shape in markets like Bangladesh, where the line between work and life was thin to begin with. A corporate culture that runs on WhatsApp and Messenger, in which a message from a senior at 11 pm carries an unspoken expectation of an immediate reply, manufactures interruption as the default and reads constant availability as loyalty. The cost is the same as anywhere, paid in shallow output and quiet exhaustion, but the cultural permission to protect one’s focus is often thinner.
| THE NOTIFICATION ECONOMY | THE FOCUS ECONOMY |
| Rewards responsiveness and visible busyness | Rewards depth and finished thinking |
| The unit of work is the reply | The unit of work is the result |
| Always on, always reachable | Protected blocks, deliberate availability |
| AI accelerates the noise | AI absorbs the shallow work |
| Produces burnout and fragmentation | Produces sustainable focus |
| Output is fast and commodity | Output is original and scarce |
The same workday, two operating systems. One rewards being reachable; the other rewards getting something hard finished.
| 23 minutes is what it takes to fully refocus after a single interruption — in a workday where one now arrives every two minutes. (Gloria Mark, UC Irvine; Microsoft, 2025) |
The collaboration counterweight
None of this means the answer is everyone retreating into headphones and refusing to reply. That would be its own failure, because collaboration genuinely competes with focus, and responsiveness exists for real reasons. A team that cannot reach one another is not deep; it is just slow. Many of the interruptions in a day are the actual work of coordinating complex, cross-functional efforts, not noise to be deleted.
The honest reading of the data is that even the always-on signals cut both ways. Microsoft’s own researchers note that evening work is not uniformly bad: for a caregiver, or a high-autonomy worker whose sharpest thinking happens at night, a boundary-less day can be a feature rather than a burden. The problem is not when people work but whether they have the agency to choose. Evening email as genuine flexibility is healthy; evening email as the overflow of an already overloaded day is corrosive.
There is a status trap here too. Deep work can curdle into a performative flex, a way of signalling importance by being unreachable, and protecting a calendar is far easier for a senior executive than for a junior whose whole job is to be responsive. A focus culture that only the powerful can access is not a strategy. It is a privilege. The real answer is design rather than withdrawal: choosing, deliberately, when a team is reachable and when it is protected, instead of defaulting to always.
Designing for depth
That design work is now a live management question. At the individual level it is unglamorous and familiar: protected blocks for deep work, notifications switched off rather than merely endured, single-tasking treated as a discipline instead of a quirk. At the organisational level it matters more, and a growing number of companies have moved on it with structural fixes such as meeting-free days, shared focus-time blocks, asynchronous updates in place of live meetings, and explicit norms about after-hours messages. The common thread is a culture that measures output rather than visible responsiveness.
AI sits on both sides of this. Used well, it absorbs the shallow work and hands people back the hours that deep thinking needs. Used badly, it does the opposite. As Microsoft’s report warns, an organisation that simply bolts AI onto its existing chaos will only accelerate a broken system, generating more messages, more meetings, and more demands at machine speed. The tool that could protect focus can just as easily destroy it. The deciding factor is whether the work around it has been redesigned.
The deep end is where humans win
The shift underneath all of this is a revaluation. For two decades the workplace rewarded availability, the person who answered fastest and was always in the channel. The notification economy took that reflex to its logical extreme and produced workforces that are busier than ever and emptier for it. The advantage now runs the other way. The scarce resource is no longer information or speed but the capacity to hold a single hard thought long enough to finish it.
The companies pulling ahead will be the ones that treat focus as the asset it has become, protecting it for their people, designing their tools and rhythms around it, and resisting the urge to fill every recovered minute with another ping. In an economy where the machines handle the shallow end, the deep end is where the humans still win.
