Work — Our approach

Work is not the opposite of rest. Fragmented work is.

The problem

Knowledge workers switch tasks on average every three minutes (Mark et al., 2005). They are interrupted — or interrupt themselves — constantly: an email notification, a Slack message, a quick check of social media, a glance at the calendar.

The cost of these switches is not just the time spent on the interruption. It is the time to recover focus afterward. After an interruption, it takes an average of 23 minutes to return to the original task at full depth (Mark et al., 2005).

This means a single interruption per hour costs roughly 38% of productive time. Most knowledge workers experience far more than one interruption per hour.

What the science says

Task-switching costs

Monsell (2003) demonstrated that switching between tasks incurs a cognitive cost: reaction time increases, error rates increase, and performance degrades. The cost is not symmetrical — switching from a simple to a complex task costs more than the reverse.

The mechanism is residual activation: when you switch away from a task, the mental model for that task remains partially active, competing with the new task’s model. This competition consumes cognitive resources.

Rubinstein et al. (2001) showed that these costs increase linearly with task complexity. For complex tasks like programming or writing, the switch cost is substantial.

Flow conditions

Csikszentmihalyi (1990) identified the conditions for flow — the optimal state of deep engagement:

• Clear goals
• Immediate feedback
• Challenge matched to skill
• No distractions

Fragmented work violates condition 4 directly. Every interruption — whether external or self-initiated — breaks the flow state. And flow takes time to enter: roughly 10-15 minutes of uninterrupted concentration (Harris, 2008).

Context-dependent memory

Smith and Vela (2001) found that memory is context-dependent: information learned in one context is harder to recall in a different one. Applied to work, this means that switching between projects — SE tickets, client calls, store management — leaves mental context residue from each project that impedes performance on the next.

This is why batching similar work together is so effective: it reduces the number of context switches and allows deep engagement with a single domain.

Job characteristics model

Hackman and Oldham (1976) proposed that meaningful work requires:

• Skill variety — using different skills
• Task identity — completing a whole, identifiable piece of work
• Task significance — understanding the impact of the work
• Autonomy — control over how to do the work
• Feedback — information about performance

Work tracking tools can support or undermine these characteristics. Tracking the wrong things (time spent, keystrokes, activity level) undermines autonomy and feedback. Tracking the right things (completed tickets, client milestones, project progress) supports them.

How Oter applies it

Multi-job separation

Oter’s Work feature supports multiple job types — SE tickets, client projects, store operations — each with its own tracking model. This reduces context-switching costs by keeping each work domain separate. When switching from client work to store work, the user changes jobs rather than mixing them in a single view.

Session-based time tracking

Rather than tracking every minute (which increases cognitive load), Oter uses session-based tracking: you start a session when you begin work and end it when you’re done. This matches the natural rhythm of focused work: a session is a unit of uninterrupted concentration.

Project and ticket hierarchies

Work items are organised into projects and tickets, creating clear task identity (per the job characteristics model). Completing a ticket is a whole, identifiable piece of work. The satisfaction of marking it done is real.

Analytics for pattern recognition

Work statistics show time spent per job, project, and ticket — but without judgment or targets. This supports the feedback dimension of the job characteristics model while maintaining autonomy. The data is for insight, not evaluation.

Practical tips

• Batch similar work. Do all your SE tickets in one block, all your client calls in another, all your store management in a third. Each block starts with a clean context.
• Use sessions, not timers. Start a session when you begin a focused work block. End it when you’re interrupted or finish. Don’t track micro-interruptions — they’re noise.
• Review your work patterns weekly. Look at what you actually spent time on. The data often reveals surprising gaps between intent and reality.
• Protect your deep work. Identify one 90-minute block per day for your most important work. Nothing interrupts this block — not email, not calls, not notifications.

References

Csikszentmihalyi, M. (1990). Flow: The Psychology of Optimal Experience. Harper & Row.

Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250–279.

Harris, J. (2008). The Cost of Interruption. Microsoft Research Technical Report.

Mark, G., Gonzalez, V. M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ‘05), 321–330.

Monsell, S. (2003). Task switching. Trends in Cognitive Sciences, 7(3), 134–140.

Rubinstein, J. S., Meyer, D. E., & Evans, J. E. (2001). Executive control of cognitive processes in task switching. Journal of Experimental Psychology: Human Perception and Performance, 27(4), 763–797.

Smith, S. M., & Vela, E. (2001). Environmental context-dependent memory: A review and meta-analysis. Psychonomic Bulletin & Review, 8(2), 203–220.