The ROI of GTD and Other Productivity Systems

We could not find a peer-reviewed study that measures how many hours Getting Things Done (GTD) or a similar system saves, so treat any fixed figure like "10 hours a week" as marketing. The return you can estimate is your own: hours recovered from fewer avoidable interruptions and forgotten tasks, minus the hours spent running the system. One week of tracking gives you both numbers.

What the Research on Switching Actually Measures

Two studies are often cited in this area, and neither supports a simple "minutes lost per interruption" figure.

• Rubinstein, Meyer and Evans (2001) ran four lab experiments in which people alternated between tasks. Switching slowed responses, the cost grew with more complex rules and shrank when people were cued about the next task. The costs discussed are in the range of fractions of a second to about a second per switch, not minutes.
• Mark, Gudith and Klocke (2008) found that people completed interrupted tasks in less time, with no difference in quality, but reported more stress, frustration, time pressure and effort.

So interruptions do have a cost, but part of it shows up as strain rather than lost hours, and the time cost depends heavily on the person and the work. That is another reason to measure your own week instead of borrowing a number.

What a System Costs to Run

David Allen's GTD method has five steps: capture, clarify, organize, reflect and engage. The "reflect" step includes a weekly review of every list and project. Time-blocking and similar methods have their own overhead. Count three costs:

• Setup: collecting and sorting everything you have open, often a full day.
• Daily upkeep: capturing and processing new items.
• Weekly review: commonly an hour or more.

How to Estimate Your Own Number

Track five normal working days and record:

• Avoidable interruptions: times you stopped to look for something, chase a task you forgot, or check something that could have waited.
• Minutes lost each time: from stopping until you were back to the same point in the original task.
• Share a system would prevent: be conservative. Meetings and urgent requests from others won't go away.

The result swings from about 67 hours to almost nothing on one assumption: how many interruptions the system actually prevents. That is why a four-week trial, followed by another week of tracking, is worth more than any published estimate.

Running a Four-Week Trial

1. Week 0, baseline: track avoidable interruptions for five working days, using a tally on paper or a note on your phone.
2. Setup: collect every open task into one list and log how long it takes.
3. Weeks 1 to 4, run it: use the system as written, and time each weekly review.
4. Week 5, measure again: track the same five days and compare with the baseline.

Net weekly gain = (baseline interruption time − new interruption time) − average weekly review time. In the example, 1.5 hours a week means about 6 hours recovered during the trial itself. If the number is close to zero or negative, drop the parts that cost the most time and keep the simplest habit that helped.

Practical Notes

• The named systems overlap. Most come down to three habits: capture tasks in one trusted place, batch similar work, and protect blocks of uninterrupted time.
• Some fixes cost nothing. Turning off non-urgent notifications has no setup or weekly overhead.
• Keep the review short. If the weekly review grows past the time it saves, the system has a negative return.

What This Doesn't Cover

This guide doesn't cover team-level causes of interruption such as meeting load, the cost of paid productivity apps, or attention difficulties that may have a medical cause. If focus problems persist despite changes to how you work, a clinician is the right person to ask.

Sources

• Rubinstein, Meyer and Evans, "Executive Control of Cognitive Processes in Task Switching," Journal of Experimental Psychology: Human Perception and Performance, 2001 (PDF)
• Mark, Gudith and Klocke, "The Cost of Interrupted Work: More Speed and Stress," CHI 2008 (PDF)