.jpg)
The message arrives at 11:14. One line, and it begins with the word quick.
Answering it takes under a minute. Finding the place again, reloading what was being held and working out what the next move was going to be takes a good deal longer than that.
Two people were involved in that transaction. Only one of them paid for it.
The transfer nobody books
An interruption is usually described as a loss, as though the time simply evaporated. It did not. It moved.
One person converted an open question of their own into a closed one. The price was met in full by somebody else, in the minutes it took to rebuild what was on the desk before the message landed.
Both parties behaved sensibly. Sending it cost the sender a few seconds and saved them an afternoon of waiting, so of course it was sent, and it would be strange to expect otherwise.
That is the shape of the condition, and it is worth being precise about it. Interruption is not a volume problem and it is not an etiquette problem. It is a transaction in which the person who spends and the person who pays are different people.
The one claim on time that leaves no record
Every other demand on a professional's week is written down somewhere, and attributable to somebody.
A deadline has a date and a person who set it, which is what the third post in this arc was about. A meeting has an organiser, a start time, an invitation, and a list of everyone else who lost the hour to it.
An interruption has none of that. Nothing anywhere records that at 11:14 one person's question ended another person's morning, and no system carries the running total.
So the cost surfaces in the only place available to it. It appears in the output of the person who absorbed it, at the point where the work does not get done.
That is where the misattribution comes from, and it is not a misreading of the evidence. It is the only evidence there is.
The explanation the market offers
A professional working inside a fragmented day is handed a familiar diagnosis.
You are distractible. You lack discipline. Turn off your notifications, put the phone in a drawer, take a course on deep work.
This publication has already made the first correction, in earlier writing on cognitive load. Fragmentation is a property of how the work has been arranged rather than a property of the person standing inside it, and the remedy for a design fault is not a wellness workshop.
The step after that is the one this post is about. Even a well arranged system contains the transfer, because the transfer is not a fault in the design. It is the ordinary operation of an arrangement in which one party spends and another party pays.
If the cost of an interruption falls entirely on the person who did not choose it, whose behaviour is the arrangement actually pricing?
The conditions that produce it
Five structural circumstances are associated with this signal, and each one describes an arrangement rather than a person.
A working environment that contains frequent demands pulling attention away from primary tasks. A meeting load that is high relative to the time available for focused work. Communication channels that create an expectation of immediate response, fragmenting concentration.
Role expectations that require frequent switching between unrelated tasks. And structural norms around availability that conflict with the conditions needed for focused work.
The last of those is not a cause in the same sense as the other four. It is a contradiction. The organisation holds two requirements at the same time, and no person can satisfy both of them in the same hour.
What tends to relieve it, and how many parties it takes
The conditions associated with relief are structural, and the interesting thing about them is how many people each one needs.
Protecting focused time explicitly in the diary rather than hoping it will appear. Grouping similar types of work within the day to reduce the cost of switching. Signalling availability and unavailability clearly, so that the expectation of constant accessibility has something to work against.
Those three describe conditions a person can produce inside the week they already have. They are not nothing, and they are the reason the individual remedies are the ones everybody has heard of.
The other two are different in kind. Agreeing response time norms that allow delayed replies to non-urgent communications. Reducing meeting frequency or duration to create structural space for uninterrupted work.
Both are agreements, and an agreement observed by one party is not an agreement. A response time norm that only the receiver keeps is not a norm at all. It is somebody being slow to reply, and it will be read that way.
Three conditions a person can build alone, then. Two that need the people generating the demand to be inside the arrangement.
The half nobody counts
The two ends of this condition look nothing like each other.
The professional sees their own fragmented day. What they cannot see is whether the fragmentation is general or particular to them, whether it has worsened across the team, or which parts of the organisation are generating it.
A leader can see that, in principle. They are the only party positioned to protect focused time across a team rather than leaving each person to defend it alone, to establish response time norms that hold because everyone is inside them, to review meeting load across the team, to set availability norms clearly, and to reduce context switching by grouping similar work where the structure allows.
The obstacle is that the organisation is counting something else. Delivery is recorded against output. Sentiment is recorded against the person.
Neither instrument holds a field for who interrupted whom, how often, or at what depth of work. A meeting is counted because somebody had to be invited to it. A message is not, because nobody has to be invited to send one.
So the arithmetic never gets done, and the first thing anyone notices is a capable person whose work has quietly got worse.
Opinion: Availability Was Never Free. It Was Never Billed.
Our position is that an interruption transfers cost from the person who initiates it to the person who receives it, and that the absence of any record of that transfer is the reason it grows.
Nothing in an organisation prices it. The sender's saving is immediate and visible to them. The receiver's cost is delayed, invisible, and attributed to the receiver.
An arrangement with those properties will expand until something else stops it, and in most organisations nothing does.
What makes this matter more than it did is the composition of the work that is left. The tasks that could be picked up and put down cheaply are the ones being automated first, and the ones that cannot are what remains.
Reconstruction cost rises with the depth of what was interrupted. A form half filled in resumes in seconds. A problem being held whole does not resume at all, it starts again.
So the price of the transfer is rising while the record of it stays at nothing. A quarter of an hour was a tolerable working unit for the sort of task that could be put down. It is not a tolerable unit for the sort that is left.
The provocation we would leave open is about accounting rather than behaviour. A meeting has a cost an organisation can state, because somebody had to be invited to it and the invitation is the record.
Nothing is created when a message is sent, which is exactly why the cost of sending one has never appeared against anybody. Most of what a working life is actually made of is spent on that basis, and an organisation that has priced only the invitations has priced the smaller half.
Declaration of Generative AI and AI-assisted technologies in the writing process:
The author made use of Generative AI or AI-assisted technologies in the preparation of this post.
The contents of this article are for informational purposes only and do not constitute professional, legal, or financial advice.
.jpg)
.jpg)
.jpg)
