You're 40 minutes into a hard problem when a coworker taps your shoulder, or your phone lights up with "call me now." You handle it. Then you sit back down and stare at your screen, unable to remember what you were even doing. That fog isn't weakness — it's a known cost called resumption lag, and there's a specific timer-based way to cut it down.

Researchers at UC Irvine found it takes an average of 23 minutes to fully return to a task after an interruption. But that number hides something important: most of that time isn't spent redoing the work, it's spent figuring out where you were. If you can shrink that re-orientation step, you can get back to real output in a fraction of the time — often 2 to 5 minutes instead of 23.

Why interruptions cost more than the interruption itself

The problem isn't the 90-second phone call. It's the "attention residue" left behind — a term coined by researcher Sophie Leroy. Part of your brain keeps chewing on the interruption (or on the task you abandoned) even after you've switched back. You're technically looking at your document again, but you're not fully there.

Unplanned interruptions are worse than planned breaks for exactly this reason: you had no chance to park your thoughts before switching. A scheduled Pomodoro break is a clean stop. A colleague asking "got a sec?" is not.

The three-timer interruption framework

The fix is to turn a messy, undefined interruption into three timed, bounded phases. Each phase has one job, and the timer's only role is to stop you from bleeding one phase into the next.

Phase 1: The 10-second capture (before you respond)

The instant you're interrupted — before you answer the question, pick up the phone, or open the email — write one line: what you were doing and the very next physical step. Not "working on report," but "writing the intro paragraph, next sentence is about Q3 revenue." This takes 10-15 seconds and it's the single highest-leverage move in this whole framework, because it's the thing that lets your brain stop rehearsing the task while you deal with the interruption.

Skip this step and you'll spend your entire return window trying to reconstruct what "capture" would have handed you for free.

Phase 2: The bounded interruption timer (2-15 minutes)

Set a timer for the interruption itself before you engage with it. A quick question gets 2 minutes. A phone call gets 10. An "urgent" email gets 15. This isn't about being rude to the person interrupting you — it's about giving yourself a hard edge so the interruption doesn't quietly expand into 45 minutes of tangential conversation, which is how a 90-second question turns into a lost afternoon.

If the timer runs out and the issue isn't resolved, say so out loud: "I've got to get back to something — can we pick this up at 3?" That sentence protects your afternoon more than any productivity app ever will.

This is where a dedicated multi-timer tool earns its keep — you need one timer counting down the interruption while your original deep work timer is paused, not stopped, so you can see exactly how much of your focus block survived. Running two timers manually with a phone clock gets messy fast, which is the actual reason people give up on this kind of structure. Download TimerForge — $17.95 for a setup that lets you run an interruption countdown and a resume-buffer timer side by side without losing track of either.

Phase 3: The 3-minute resume buffer

This is the phase almost everyone skips, and it's the one that eliminates most of that 23-minute lag. Before diving back into the actual work, set a short timer — 3 minutes is enough — and use it only to re-read your capture note, glance at the last few lines you wrote or the last file you had open, and re-state your next physical step out loud or in writing.

Do not use this buffer to check "just one more thing." That's the trap: the resume buffer is reconstruction time, not a second interruption window. Its entire purpose is to let your brain re-load context before you ask it to produce anything, which is exactly what separates people who bounce back in 5 minutes from people who lose the rest of the hour.

What this looks like in practice

Say you're deep in a coding session and Slack pings with "quick question about the API." Capture: "refactoring the auth middleware, next step is testing the token refresh path." Interruption timer: 5 minutes for the Slack thread. Resume buffer: 3 minutes to reopen the file, re-read your capture note, and glance at the failing test. Total overhead: 8 minutes of structured handling instead of an open-ended derailment that could easily eat 30.

The framework also works for interruptions you generate yourself — the urge to check email mid-task counts too. Capture your place, give the urge its own 2-minute timer if it's genuinely urgent, then run the resume buffer before returning to work.

Common mistakes that break the framework

The most frequent failure is skipping Phase 1 because the interruption "seems quick." It's precisely the interruptions that feel quick that end up costing the most, because you never captured your place and have nothing to reconstruct from. The second most common mistake is treating the resume buffer as optional — people dive straight back into work "to save time" and end up producing 10 minutes of unfocused, low-quality output before actually re-engaging, which takes longer than the buffer would have.

One thing to do today

Next time you're interrupted mid-task, pause for 10 seconds and write down exactly what you were doing and your next concrete step before you respond to anything. That single habit, done consistently, will cut your recovery time more than any other change on this list — and it costs you nothing but ten seconds.