You’re about to start your most important work of the day. Your calendar is blocked, your coffee is fresh, and you’ve finally carved out two uninterrupted hours. Then you remember: you need to send that quick email. It’ll take two minutes, you tell yourself. Might as well knock it out now.
Forty-five minutes later, you’re still not working on your important task. The “quick” email required you to find an attachment. Which reminded you to update that spreadsheet. Which made you notice a Slack message that needed a response. Which led to a brief conversation that surfaced another small task. Now your blocked time is half gone, and you haven’t even started.
This isn’t a discipline problem. It’s a design problem.
The Mechanism: Why ‘Just One More’ Never Stays Small
When you tell yourself a task will take “just two minutes,” you’re making a prediction about task duration. And humans are reliably terrible at this specific type of prediction.
The planning fallacy—a term coined by Daniel Kahneman and Amos Tversky—describes our systematic tendency to underestimate how long tasks will take. But with “just one more thing,” something else is happening too: you’re not accounting for task switching costs.
Sophie Leroy’s research on “attention residue” shows that when you switch tasks, part of your attention remains stuck on the previous task. The effect intensifies when you switch to a task but don’t finish it, or when the task you abandoned was important to you. Your brain doesn’t cleanly toggle between contexts—it drags the previous context along, reducing your cognitive capacity for what comes next.
The Hidden Expansion Pattern
Here’s what actually happens when you add “just one more thing”:
The task itself takes longer than expected (planning fallacy). Then you encounter friction—finding a file, waiting for a page to load, remembering a password. This friction creates small pockets of time where your mind wanders to related tasks. As David Allen describes in Getting Things Done, these become “open loops”—incomplete commitments that consume background processing power.
Even after you finish, there’s a cognitive switching penalty. Your brain needs time to reload the context of your original task. The studies vary, but the switching cost ranges from several seconds to over twenty minutes depending on task complexity. Not the time spent on the interrupting task—the time to fully re-engage with what you were doing before.
“Just one more thing” compounds this effect. Each small addition creates switching costs in both directions—away from your main work and back to it.
Why We Keep Doing It Anyway
If the costs are so high, why does “just one more thing” feel so compelling?
First, small tasks offer what behavioral economists call “completion utility”—the psychological reward of checking something off. Your brain treats completion as inherently valuable, independent of the task’s actual importance. This is the Zeigarnik effect in reverse: incomplete tasks create tension, and our minds are drawn to relieving that tension through completion.
Second, important work often involves uncertainty and discomfort. Starting that complex project means confronting questions you don’t have answers to yet. The “quick” email offers certainty—you know exactly what to do and how long it should take. It’s procrastination wearing the costume of productivity.
Third, we’re notoriously bad at protecting future time. When you tell yourself “I’ll just do this now and start my deep work after,” you’re making a prediction about Future You’s circumstances. But you’re making that prediction with Present You’s optimism bias firmly in place. Gerd Gigerenzer’s work on risk perception shows we systematically underweight delayed costs compared to immediate rewards—even when the delayed cost arrives in twenty minutes.
What Actually Works
1. Use an ‘Interrupt Capture’ System
Keep a designated place—digital or physical—where you immediately record any task that occurs to you during protected time. Don’t evaluate it, don’t “just quickly check,” don’t decide if it’s urgent. Write it down and return to your work.
The mechanism: This leverages the Zeigarnik effect productively. Your brain needs assurance the task won’t be lost. A reliable capture system provides that assurance without requiring task completion. You’re not ignoring the task; you’re deferring the decision about when to do it.
2. Implement Minimum Task Batching
Decide in advance that you never do fewer than three email-sized tasks in a single session. If you’re going to pay the switching cost of moving into “small task mode,” extract value from it by handling multiple items before switching back.
The mechanism: This amortizes your switching costs across multiple completions. Instead of paying the cognitive penalty four separate times for four small tasks, you pay it twice—once to enter small-task mode, once to exit. The math is simple: fewer switches means lower total cognitive cost.
3. Build Specific Re-engagement Protocols
Before you do anything else—even legitimate interruptions—write one sentence describing exactly where you are in your main task. Not just “working on report” but “deciding whether the third example supports the main argument or needs replacement.”
The mechanism: This creates a cognitive bookmark. When you return, you’re not just remembering what you were working on; you’re remembering your exact thought process. You’re reducing the re-engagement cost by preserving your mental state. Cal Newport describes this as “attention engineering”—deliberately designing aids that make deep work resumption easier.
4. Deploy Pre-commitment Through Time Constraints
Set a timer for your deep work blocks, but here’s the critical part: the timer doesn’t tell you when to stop. It tells you the earliest point at which you’re allowed to check if anything urgent has emerged. Before the timer, the answer is automatically no.
The mechanism: This removes the decision point. “Should I check this now?” is a negotiation your present self will lose because you’re already thinking about the other task. A pre-commitment removes the negotiation. It’s Odysseus and the sirens—you’re binding yourself to the mast before you hear the song.
5. Audit Your ‘Just One More’ Accuracy
For one week, every time you think “this will just take two minutes,” write down your prediction. Then track the actual time from start to full re-engagement with your original task.
The mechanism: You’re building calibration. Most people discover their “two-minute tasks” consistently take fifteen to thirty minutes when switching costs are included. This data defeats your optimism bias. Next time the urge strikes, you’ll have evidence, not estimation.
The Reframe: Treat Transitions as Infrastructure
You don’t have a discipline problem when you add “just one more thing.” You have an infrastructure problem. Specifically, you’re trying to operate without transition systems.
Knowledge workers treat task switching like it’s free—or at least cheap enough to be invisible. But attention is infrastructure, like roads or electricity. You wouldn’t constantly turn your computer off and on throughout the day, even though each restart only takes a minute. The cumulative cost would be absurd.
The same logic applies to your attention. Each “just one more thing” is a restart. And just like your computer, your brain works best with fewer restarts and longer periods of sustained operation.
This isn’t about perfect focus or monk-like discipline. It’s about honest accounting. When you consider adding one more small task, you’re not deciding whether to spend two minutes. You’re deciding whether to spend two minutes plus two switching penalties plus the planning fallacy margin. That’s frequently a fifteen-to-thirty-minute decision.
Sometimes that’s worth it! Urgent truly is sometimes urgent. But most of the time, you’re trading a small reduction in task-list anxiety for a large reduction in deep work capacity.
Try This Week
Pick your single most important work block this week. Before it starts, write down three specific tasks you will not do during that block, even if you think of them. Put them in your capture system instead.
The goal isn’t to prove you have willpower. It’s to test whether protecting one block of time yields noticeably better work than your usual approach. Run the experiment, then decide if the trade-off is worth making permanent.
