There's a statistic that gets cited in every productivity article, every remote-work deck, every manager training on focus and flow. You've probably seen it. You may have even used it
It takes 23 minutes and 15 seconds to fully refocus after an interruption.
The number comes from Gloria Mark, a researcher at UC Irvine, who has spent decades studying how knowledge workers actually spend their time. In a Gallup Business Journal interview — confirmed in a Fast Company interview that has since been cited approximately everywhere — she described what her field studies found: when workers get interrupted, they don't just snap back. They take two detours first. And by the time they get back to the original work, an average of 23 minutes has passed.
That number traveled fast. It became the foundation of arguments for async-first work, for notification batching, for deep work blocks, for every "protect your focus time" framework that anyone has ever written a LinkedIn post about.
Here's what nobody mentions about that research: Gloria Mark was studying office workers 🏢. Knowledge workers in open-plan environments, dealing with email and instant messages and colleagues stopping by their desks. There was no neurodivergent-specific analysis. There was no remote-specific analysis. There was no ADHD cohort.
The 23-minute rule describes how a neurotypical brain in an office environment recovers from interruption.
And then we took that number and applied it to everyone.
Mechanism 01
Task-switching is not automatic.
For an ADHD brain, the friction of reorienting after interruption is structurally heavier. Research published in PubMed found that children with ADHD showed "substantially larger switch costs" than non-ADHD controls — the cognitive effort required to reorient is meaningfully greater. Reorienting requires active executive effort. And executive effort is exactly the resource that ADHD taxes most heavily 🧠.
Dr. Thomas E. Brown, a clinical psychologist who has spent his career studying ADHD, describes what interruption feels like from the inside: "losing the thread." Thoughts scatter. A sentence that was forming mid-air vanishes. The entire internal direction of the work is gone — not paused, gone.
Mechanism 02
Time blindness means the interruption feels unbounded.
A 2021 review published in Frontiers in Psychiatry identified time perception deficits as a central feature of adult ADHD — not a secondary nuisance — and argued they may be "at the very root of ADHD-related symptoms."
When a neurotypical worker gets interrupted, their internal clock says: five minutes have passed, get back. For an ADHD worker, that clock doesn't update reliably. "A few minutes" can feel the same as 45 minutes. The interruption expands to fill available space ⏰.
Mechanism 03
Emotional dysregulation extends the recovery window.
Emotional dysregulation is a core feature of ADHD, not a side effect. When an interruption breaks a state of deep concentration, the irritation, the sense of loss, the spike of anxiety — these don't fade in a few minutes. That emotional layer means you're not just reconstructing where you were. You're managing a frustration response before cognitive recovery can even begin
I see this with my son. When he’s in deep-work mode and mom or dad need his help around the house, it’s not just a simple break. It’s not a reply that the time isn’t right. It’s an emotional experience and frustration for him. Not just for the temporary redirection but the time and effort required he knows it will take to get back into the groove.
Put the three mechanisms together, and you get a refocus window that is longer than 23 minutes — potentially significantly longer — for a meaningful percentage of your workforce.
By how much? No one knows. Because no one has measured it.
The Research Gap Is the Story
Here's what I find remarkable: the 23-minute number has been cited in productivity contexts for nearly two decades. It has driven policy decisions. It has shaped workplace design. It has been used to justify async-first communication norms, notification batching protocols, open-office redesigns.
And in all that time, no researcher has run a controlled study on what interruption recovery looks like specifically for ADHD adults in knowledge work environments.
There is clinical research on task-switching costs in ADHD. There is research on time perception deficits. There is research on emotional dysregulation. There is no equivalent of Gloria Mark's field study — naturalistic, workplace-based, longitudinal — with an ADHD-specific cohort.
We have a number that every productivity researcher, remote work advocate, and workplace designer has been citing as a universal baseline. And we have a population that represents somewhere between 5 and 15 percent of the current adult workforce, for whom every single mechanism that drives that number is physiologically amplified.
The finding
Nobody has measured what the refocus time actually is for ADHD workers. 📊
That is not a footnote. That is the finding.
What Remote-Native Companies Got Right (By Accident)
GitLab — one of the largest all-remote companies in the world, with over 1,600 employees across 65+ countries — built its communication culture around a core principle: asynchronous by default, real-time by exception.
Doist — the company behind Todoist and Twist — has operated on the same principle since day one. In an interview with Slab, their content lead Fadeke Adegbuyi put it plainly: "When we say that we're async first, what we mean is that we're writing first."
Here's what these companies understood, even if they weren't thinking about neurodiversity when they designed their norms: if the message doesn't interrupt, there's no interruption to recover from.
For neurotypical workers, async-first is a nice-to-have. For ADHD workers, it is structural protection. It is the difference between losing 23 minutes and losing an hour and a half — or losing the entire afternoon to an escalating frustration spiral that started with one "quick question" Slack message at 2pm 🌐.
What This Means for the Office
You may be reading this and thinking: we're not a remote company. What does any of this have to do with us?
A lot, actually. The underlying principle — interrupt less, recover more — applies everywhere.
🎧 The headphone norm, done right.
Make it explicit: headphones + a simple desk card (red = focus mode, green = available). It takes 30 seconds to introduce. It eliminates the constant ambiguity of "is it okay to interrupt this person right now" — which benefits everyone.
🗓 Office hours for "quick questions."
Instead of continuous availability, create a defined window when you are available. For ADHD workers who struggle with "now vs. not now" time perception, externalized structure like office hours gives a clear fence around focus blocks, which makes them significantly easier to maintain.
💬 Kill the walk-by — with a system, not a social norm.
The fix isn't telling people not to interrupt each other — that never works. The fix is a lightweight async channel specifically for non-urgent questions, and making it the default first step before walking to someone's desk.
The Invitation in the Data Gap
We don't have the data on ADHD refocus times. The research doesn't exist yet. And I'm not sure any of us — researchers included — know what the number actually is.
But some of you do.
If you have ADHD and you work in a knowledge work environment, you have lived this data. You know whether 23 minutes is laughable, aspirational, or actually close to your experience. You know what the emotional layer costs you, how long before you can actually reconstruct where you were.
That knowledge is not anecdote. That is primary data that nobody has collected yet.
How long does it actually take you to get back?
Not the 23-minute average. Not the neurotypical baseline. Your number, in your work environmen. Drop it in the comments, or reply directly to this email. I'm genuinely collecting these responses, and I'll write about what I hear.
What to Do With This
Remote leaders
Look how to pivot towards async for your communication stack. Audit your async norms. If your Slack culture expects same-hour responses— you have a continuous interruption environment with a documentation layer on top. If you need help on those 1st steps towards async, I did a podcast all about this.
Office + hybrid leaders
Introduce a visible focus signal that has teeth — a shared norm the whole team agrees to honor. Like a colored card on everyone’s desk. Green = talk to me. Red = Go Away! Start with a 30-day experiment. Track what works.
ADHD workers
You are allowed to protect your focus time. The accommodation doesn't have to be medical or formal — it can just be a conversation about how you do your best work. Start there.
Everyone
The 23-minute rule was always an average, in a specific population, in a specific environment, at a specific moment in time. It was never a universal law. Ask what the actual number is for you.
The data gap on ADHD and interruption recovery is not just an academic problem. It's a design problem. Every workplace that optimizes for the 23-minute baseline and never asks whether that baseline applies to their entire workforce is making a structural choice, whether they know it or not.
The good news is that the structural choice can go in the other direction just as easily.
I'd love to hear what brought you here. Reply directly — I read every message.
If this resonates, share it with a CEO or HR leader in your network. The conversation needs to start somewhere.
And if you're past the conversation stage and ready to actually build the systems — reach out. That's exactly what I do.