The measured cost of context switching, applied to subscription apps

Interruption cost is among the most replicated findings in HCI research.

Mark, Gudith & Klocke's 2008 ACM CHI paper is the most cited empirical study of workplace interruption. The team observed information workers in situ and measured the cost of switching between tasks under varying interruption conditions.

Mark, Gudith & Klocke: "When people are interrupted, they take an average of 23 minutes and 15 seconds to return to the original task… Interruptions impose a longer return time when the interrupted task is more difficult." — Mark, G., Gudith, D., & Klocke, U. (2008). "The Cost of Interrupted Work: More Speed and Stress." Proceedings of the 2008 CHI Conference on Human Factors in Computing Systems, 107–110.

The earlier theoretical foundation is Monsell's review of task-switching costs in Trends in Cognitive Sciences. The mechanism Monsell described — a residual "switch cost" that persists even after the new task has begun — is what makes interruption cumulatively expensive.

Monsell: "It typically takes longer, and one is more error-prone, when, on each successive trial in a sequence, one must switch task… These costs are evidence of the time required for a control mechanism to adjust to the new task set." — Monsell, S. (2003). "Task Switching." Trends in Cognitive Sciences, 7(3), 134–140.

For subscription apps specifically, each app is a separate task context: distinct UI, notification stream, mental category. The Mark et al. finding generalizes — each switch between subscription apps carries a return-time cost proportional to the task's difficulty.

The intervention with the most consistent support: batching. Group app usage into deliberate windows rather than letting switches happen continuously. Reducing the total number of subscription apps with persistent permissions reduces the surface area of potential switches.

References

• Mark, G., Gudith, D., & Klocke, U. (2008). ACM CHI, 107–110.
• Monsell, S. (2003). Trends in Cognitive Sciences, 7(3), 134–140.

Related: Notification fatigue · Attention economy · Subscription fatigue