Protect a block of time, remove the things that interrupt you, and expect the benefit to show up as more words and less stress rather than as speed.
That last point is the opposite of what you have been told, and it comes from the research the popular advice cites without reading.
The Famous Number Does Not Exist
You have seen this everywhere: it takes 23 minutes and 15 seconds to get back on task after an interruption. It is always attributed to Gloria Mark's research at the University of California, Irvine.
Searching the full text of both papers, the strings "23 min" and "15 sec" appear nowhere in either. The figure is second-hand and does not match its cited source.
The actual measured number, from Mark, Gonzalez and Harris, "No Task Left Behind? Examining the Nature of Fragmented Work" (CHI 2005):
"When people did resume work on the same day, it took an average length of time of 25 min. 26 sec (sd=54 min. 48 sec.). This may seem like a relatively short amount of time, but it is also important to consider that before resuming work, our informants worked in an average of 2.26 (sd=2.79) working spheres."
Two things about that are more important than the digits.
The standard deviation is 54 minutes 48 seconds, more than double the mean. The distribution is enormously skewed, so a single average badly misrepresents it. Some resumptions are almost immediate and some take hours.
It is not refocus time. It measures elapsed time before returning to a task, during which the person did an average of 2.26 other things. The popular version misdescribes what was measured, not just the number.
The study itself was solid: 24 information workers shadowed in person with a stopwatch, "Over 700 formal hours of observation" across thirteen months. They were information workers rather than writers, which is worth knowing before generalising.
The Real Finding Is Better Than the Myth
Mark, Gudith and Klocke's follow-up, "The Cost of Interrupted Work: More Speed and Stress" (CHI 2008), with 48 subjects, found something genuinely surprising:
"Surprisingly our results show that interrupted work is performed faster. We offer an interpretation. When people are constantly interrupted, they develop a mode of working faster (and writing less) to compensate for the time they know they will lose by being interrupted. Yet working faster with interruptions has its cost: people in the interrupted conditions experienced a higher workload, more stress, higher frustration, more time pressure, and effort."
Read that again in a novelist's terms. Interruption did not make people slower. It made them work faster and produce less text, while feeling significantly worse. The measured word counts bear it out: output was longest in the uninterrupted condition.
And the threshold is low:
"After only 20 minutes of interrupted performance people reported significantly higher stress, frustration, workload, effort, and pressure."
Twenty minutes. Not a day of fragmentation; one short stretch.
So the honest case for distraction-free writing is not that interruptions cost you time. It is that they cost you words and wellbeing, which for a novelist is exactly the loss that matters.
The Cruel Finding
One more line from the 2005 paper, which argues for protecting deep work specifically:
"The irony of the work day is that the longer people spend in a working sphere, (and thus we assume become more involved in it) the more likely it is to be interrupted and the longer is the interrupting event."
The deeper you are in, the more likely you are to be pulled out, and the longer the pull lasts. Absorbed writing is the most vulnerable state, not the most protected one. That is an argument for defending the block in advance rather than relying on concentration to hold.
The same study found "57% of their working spheres are interrupted", with an average of just "11 min. 4 sec." in a task before switching or being interrupted.
What Actually Helps
The research supports the principle rather than any particular technique, so treat what follows accordingly.
Remove the interruption source, not your response to it. Self-interruptions resumed more slowly in the 2005 data, at 29 minutes against 22 for external ones. Willpower is the weaker lever; closing the tab is the stronger one.
Protect a block, and tell people. Since long absorbed stretches attract interruption, the defence has to be arranged before you start.
Judge the session by words, not minutes. Interrupted work felt faster and produced less. Time at the desk is the misleading measure.
Expect the benefit to be felt as calm. Lower stress, frustration and effort were the measured differences, and they are what makes a daily practice sustainable.
What Deckle Does
Deckle has a Focus Mode, which strips the interface back to the text, and a typewriter scrolling option that keeps the current line centred. Both remove in-application distraction, which is the part a writing app can control.
Two features connect to the findings above.
The sprint timer tracks active writing minutes, recorded as you actually type rather than as elapsed wall-clock time. Since the research says interrupted sessions feel faster while producing less, a measure of real writing time is more honest than a stopwatch.
Word tracking is net rather than gross. Deleting text takes back the words it added, floored so a day cannot go negative. That is deliberate: it measures how much closer the manuscript got, not how much typing happened. Given that interrupted work produces less text while feeling faster, a net measure is the one that reflects the finding.
For the day-to-day practice, see writing distraction free in Deckle, and for the surrounding habit, building a daily writing habit.
Common Mistakes
- Quoting the 23 minute figure. It is not in the research it is attributed to.
- Treating that number as refocus time. The real measure is elapsed time before returning, with other tasks in between.
- Assuming interruptions make you slower. Measured result: faster, with less output and more stress.
- Relying on willpower. Self-interruptions took longer to recover from than external ones.
- Judging a session by time spent. Interrupted work feels faster and yields fewer words.
- Generalising these studies to novelists without saying so. They studied information workers and students.
FAQ
Q: How long does it really take to refocus after an interruption? A: The cited study measured 25 minutes 26 seconds to return to a task, with a standard deviation of nearly 55 minutes. It did not measure refocus.
Q: Where does "23 minutes 15 seconds" come from? A: Not from Mark's papers. The strings do not appear in either full text.
Q: Do interruptions make me slower? A: No. The measured effect is that you work faster, write less, and feel considerably worse.
Q: How long before interruption starts hurting? A: Significantly higher stress, frustration and effort were reported after only 20 minutes of interrupted work.
Q: Does any research cover novelists specifically? A: Not that I could find. These studies used information workers and university students.