Focus & Productivity · Real working days
How to Get Real Work Done When Your Day Is Full of Interruptions
Some days you can't turn notifications off. People genuinely need you. Here's a way to protect meaningful work without pretending you control every hour.
By eleven in the morning you've answered three questions that weren't on your list, joined a call nobody planned, helped somebody find a document and started the same report four times. The report is still two sentences long. And you're not even sure you had a bad morning: you helped people who needed help.
Most concentration advice is written for somebody who can shut the door and silence the world. But support teams, teachers, parents, carers, coordinators and people in operational roles don't necessarily have that choice. Sometimes being available is the work. So the right question is not how to become unreachable. It's how to make progress between genuine demands without carrying ten half-finished thoughts around all day.
The first adjustment
Make two lists, not one: work that must be responsive and work that benefits from continuity. Decide how people reach you for genuine urgency. Then give the continuity work smaller, returnable stages, each with a written next action. You're designing for interruptions that will happen, not praying that they won't.
Your day may have two different jobs inside it
Consider somebody in customer support who is responsible for answering incoming requests and improving the help centre. The requests are interrupt-driven. The documentation improvement needs an uninterrupted stretch. If the calendar treats both kinds of work as interchangeable, the second one usually loses.
There's nothing inherently wrong with that on a day when customers urgently need help. But it matters to name the trade-off. If management wants a documentation update and uninterrupted support coverage from the same person in the same hour, the missing ingredient may be capacity, not personal discipline.
Make a simple agreement about availability. Who is on primary coverage? Who is backup? What counts as urgent enough to interrupt focused work? When can non-urgent questions wait? If you're working alone or caring for someone, the agreement may be with yourself and the people around you rather than a team.
Find the difference between an interruption and an emergency
Everything arriving unexpectedly can feel urgent in the moment. That isn't the same as everything needing an immediate answer. A broken production service, a safety concern or a dependent person's genuine need should be handled promptly. A routine request for a link may be able to wait fifteen minutes.
Build a short triage rule: now for actual incidents and agreed urgent needs, next check for important but non-urgent work, and scheduled for work that doesn't require immediate attention. Your team, organisation or caregiving circumstances determine what these labels mean. An article can't invent that policy for you.
A practical interruption agreement
Urgent: Call or use the established incident channel. A designated person responds.
Routine questions: Put them in the usual queue. They'll be reviewed at the next agreed checkpoint.
Collaborative work: Include context, the decision needed and a deadline so another person can help asynchronously.
Private emergencies: Follow the appropriate safety or support procedure rather than relying on a productivity tool.
Don't publish a message saying you're checking email at 4 p.m. if your role requires you to respond to customers all day. The good plan is the one that fits the actual responsibilities, not the plan with the prettiest blocks.
Leave a landing strip for your future self
A useful way to make interruptible work less painful is a tiny “ready-to-resume” note. Before switching away, record exactly where you are, what you were deciding, and the next move. Not a general reminder such as “finish report”, but something like “The introduction is done; verify the March figure in the finance sheet, then write the second comparison paragraph.”
There is research behind this idea. In a 2018 Organization Science paper, Sophie Leroy and Theresa Glomb studied how anticipation of returning to unfinished work can affect performance on an interrupting task. Briefly planning the return helped in their experiments. It's not a guarantee that a sticky note solves every interruption, but it is a reasonable habit to borrow.
The thirty-second return note
I was doing: Comparing two suppliers' terms.
Finished: Price and delivery sections checked.
Still uncertain: Whether support is included in quote B.
Next exact action: Open supplier B's agreement at clause 4 and check inclusions.
Where it lives: Link to the comparison document.
On return you don't need to reconstruct the entire mental state from scratch. You can begin with a verb. That matters when interruptions aren't occasional defects but part of the day's operating conditions.
Build work in pieces that leave something behind
Deep work is valuable, but not every useful deliverable requires an uninterrupted ninety-minute performance. Break the report into a factual outline, one argument, a data check, a review and a final edit. Each piece should leave an artefact another session can pick up.
This isn't the same as fragmenting every task into dozens of microscopic chores. The pieces need to correspond to real progress. A list saying “open document, choose font, write heading” can make the planner feel busy while the thinking stays untouched.
Try a simple test: if you're interrupted after twenty minutes, could someone else tell what you accomplished? If not, perhaps the unit of work is too vague. “Research three candidates with sources” is a firmer stopping point than “work on project”.
Why time blocks can still help, but differently
A fixed two-hour block might be impossible for somebody responsible for a live queue. A thirty-minute opportunity window with a preselected task may be realistic. The distinction is subtle: the block is a chance to focus, not a promise the world won't knock.
Put work that has a real deadline and requires thought into your quieter predictable periods where possible. Put reactive tasks into coverage windows. When traffic spikes, don't treat the protected period as a personal failure; mark what actually happened and talk about staffing if the pattern repeats.
When you're doing household or care work, predictability may come in smaller pieces. “After the school run” and “when someone else is home” can be more useful planning units than precise clock times.
A worked example: the service desk with a writing deadline
Illustrative case, not a reported workplace outcome. Leah handles user tickets and has promised a new help article by Friday. On Wednesday the queue is unusually busy. She can't decline customer work, but she can change how the draft is organised.
She agrees with her lead that a colleague will cover non-urgent new tickets for one short slot on Thursday if volume permits. She prepares a factual outline today, leaving sources and missing questions beside each section. On Thursday she uses the slot to draft the complicated explanation first. The article gets reviewed in two short rounds.
If the team cannot provide coverage, the deadline needs renegotiating. Leah should not secretly work until midnight to make an impossible plan look possible. A fair process produces either capacity or a changed expectation.
Make escalation easier than anxious checking
People check messages compulsively when they don't trust the route for urgency. If an important issue could arrive anywhere, you feel obliged to watch everywhere. That is an organisation problem.
Where your role allows it, agree on one urgent route and a separate ordinary queue. Make sure colleagues know the difference and that the urgent route is monitored. If you're not on duty, an agreed backup should exist for critical matters. Do not substitute a focus timer for an on-call schedule.
Then practise closing the other surfaces during a bounded work session. Not because they're unimportant, but because the urgent work has a reliable home rather than being mixed indiscriminately with reminders, newsletters and social conversations.
The difference between being interrupted and switching voluntarily
Some interruptions arrive from other people. Others are self-initiated: you open messages because you're unsure how to continue, or browse the help centre for a fact you already know. These deserve different responses.
If someone interrupts, the ready-to-resume note can reduce the restart cost. If you keep checking voluntarily, ask what happens just before you switch: uncertainty, boredom, waiting for a result, anxiety about an unanswered thread? Maybe you need a smaller next action rather than stricter blocking software.
The context switching guide covers voluntary shifts in more detail. Here the emphasis is on a day where competing demands are real and sometimes unavoidable.
Try one honest daily capacity calculation
A day that actually contains people
Eight hours available does not mean eight hours of project work. Imagine three hours of scheduled coverage, two hours of meetings, an hour of follow-up and ordinary rest breaks. You may have only a few useful pockets left, and a surprise incident could remove them.
Write down the fixed obligations before planning individual work. Use the remaining time for one or two real deliverables, not a fantasy set of seven. If the numbers don't add up, flag the conflict rather than hiding it inside your evening.
This is not a mathematical model for every profession. Some shifts genuinely require continuous live availability; others have more control. But a rough honest view of capacity is much better than assuming all unoccupied squares belong to the project you want to do.
When home responsibilities are the interruptions
Children, elderly relatives, pets and shared households create interruptions that don't always fit an incident queue. Start with realistic agreements rather than treating loved ones as distractions to be suppressed. A parent supervising a child cannot safely promise uninterrupted work because a timer says so.
Can another adult cover for a known period? Can a demanding task be shifted to a quieter hour? Can the employer agree flexible delivery times? Can some work be done offline while remaining available for household needs? These depend on your circumstances, and you don't owe the internet your family's private details.
There are days when the right answer is less work or outside help, not a more determined study soundtrack.
What should you tell colleagues when an interruption keeps recurring?
Three messages that don't make anybody the villain
For a non-urgent request: “I can look at that after the current customer issue. Could you add the link and when you need the answer?”
For coverage conflicts: “The queue has been taking the time reserved for the documentation work each day. Can we agree a coverage slot or move the documentation date?”
For a complex interruption: “This needs a decision rather than a quick answer. Could we put the options and owner in the project note so the next person has the context?”
None of these scripts promises an instant response. They make the work visible and help the recipient know what happens next. That's kinder than disappearing and kinder than saying yes to everything with no feasible way to finish.
A one-week experiment for an interruption-heavy role
For five workdays, pick one continuity task and leave a ready-to-resume note every time you switch away. Record the actual interrupts in broad categories, not personal surveillance: customer need, scheduled meeting, household, self-switch, unexpected operational issue.
At week's end, don't grade yourself on how long you remained uninterrupted. Ask whether you resumed more easily, whether the important work moved, and whether genuine urgent demands still got timely attention. If interruptions are consistently swallowing the entire week, use the evidence to change staffing or priorities.
Research on attention residue is useful, but it doesn't replace organisational design. A UK HSE reminder about breaks and changes of activity is also worth noting for screen-heavy roles: people need real variation and rest, not a sequence of intense tasks with no recovery.
A small Calphii way to make returning easier
We built Calphii as a personal place to keep notes, small tasks, ambience and Deep Focus together. If you're taking a short concentration opportunity, leave the precise return instruction in a note and put the current outcome on your board. When the session ends or someone needs you, you can resume without rebuilding the whole plan.
Calphii isn't an emergency alert system or workforce scheduler. Your real response obligations come first. For estimating realistic work windows, read our guide to task estimates; for the emotional strain of a fragmented day, the overwhelm guide may also offer a useful starting point.
The day you couldn't control can still count
Maybe you'll finish the report. Maybe you'll only move it to a clear outline and handle six urgent problems. Either can be a legitimate day's work. A better system helps you see what happened and protects a way back to your priorities without demanding that real people stop needing you.
When you plan for interruptions instead of treating them as moral failures, the work gets a little more manageable. So do the days around it.