Taming Slack and email

It is 10:43 a.m. on a Tuesday. You are fifteen minutes into a problem you have been trying to sit with for three days. You have the tab open, the notes in front of you, the first real thread of clarity forming. Then Slack lights up. Then another. Your phone buzzes. An email preview appears at the top of your screen. You glance — just for a second — and the thread dissolves. By the time you close the notifications and try to find your way back, the problem has gone cold again.

This happens to knowledge workers dozens of times a day. Most treat it as the weather: unfortunate, unavoidable, worth complaining about but not worth changing. It is not unavoidable. It is a system you have accidentally opted into. And like any system, you can redesign it.

What notifications actually are

A notification is a request to switch context. It is not inherently urgent. It is not even inherently meant for you right now. It is an asynchronous message that someone sent at a time convenient for them, arriving on a channel that has been misconfigured to demand your immediate attention.

The platform is not neutral here. Slack’s default is real-time. Email clients badge aggressively. Both systems treat every incoming message as an event worth interrupting you for, because engagement is the metric that matters to the software, not your thinking time.

The result is a workday that feels like an air traffic control tower: inputs arriving from every direction, your job reduced to routing responses. You are always reacting. The deep work — the analysis, the creative problem, the hard decision — gets pushed to the margins.

The fix requires you to do something the platform does not encourage: opt out of real-time.

The batching principle

Batching means designating two or three windows per day for reading and responding to messages — and not checking outside them.

A common structure for an eight-hour workday: one window at the start of the day (to catch overnight threads and set the day’s direction), one at midday (to unblock colleagues and handle anything time-sensitive), one in the final thirty minutes (to clear the queue before closing). Everything between those windows is off-limits.

This is not a radical idea. It is how email worked before smartphones made it perpetually ambient. The difference is that you have to enforce it intentionally now, because the defaults work against you.

The argument against batching is usually urgency. “But what if something important comes in?” The honest answer: almost nothing in most knowledge-work environments is so urgent that a two-hour response window will cause meaningful harm. The things that are genuinely that urgent — a production incident, a client in crisis, a colleague blocked on a critical path — are not things people resolve through Slack messages. They call. They escalate. They find you.

What looks urgent in a notification is usually just new. Novelty and urgency are not the same thing.

Synchronous urgency versus manufactured urgency

Synchronous communication — a call, a live meeting, someone walking over — has a built-in cost signal. The person initiating it has to arrange it, show up, and give you their full attention. That cost filters for genuine need.

Asynchronous channels like Slack and email have near-zero cost to send. Someone can ping you while doing three other things, for reasons ranging from critical to “I could have found this myself in thirty seconds.” The friction of receiving the message is borne entirely by you.

Manufactured urgency is what happens when a low-stakes message arrives on a high-urgency channel. The person who sent it probably was not trying to manipulate you. They were using the fastest tool available, because the cost to them was zero. But you absorb the interrupt anyway.

The practical test: ask whether this conversation would justify pulling someone into a meeting right now. If the answer is no, it can wait for your next message window. If the answer is yes, pick up the phone. Stop trying to simulate real-time conversation through a text channel. The hybrid is worse than either extreme.

Turning off notifications without becoming unreachable

The objection here is reachability. Colleagues will wonder where you are. Your manager will notice you are not responding in the usual rhythm. Projects will seem to stall.

Three things counteract this: your status, your norms, and your responsiveness when you do show up.

Status. Most teams underuse Slack’s status field. Set it explicitly when you are in a focus block: “Deep work until noon.” Update it when you are available. A visible status is a passive signal that explains your absence without requiring you to announce yourself. It converts what might read as ignoring into a legible working style.

Norms. If you are in a position to influence how your team communicates — and even junior contributors often are, by modeling behavior — be explicit about response time expectations. “I check Slack at 9, 12, and 4” is not a policy document; it is a sentence you can put in your bio or mention in your next one-on-one. Teams normalize whatever their highest-responders do. If you are currently fastest to reply, your speed has become the floor others feel pressure to match. Slowing down, with explanation, often gives colleagues permission to do the same.

Responsiveness when present. The trade-off for batching is that your windows need to actually be good. When you are in a message window, you are fully in it: reading carefully, giving complete answers, clearing blockers for others. If your batched responses are just as shallow as your real-time ones, you have not improved anything. Doing fewer things well is the point.

The interrupted day versus the batched day

The diagram below shows what happens to your attention across a workday in each mode. The interrupted day is not just slower — it structurally prevents the kind of focus that hard problems require.

The interrupted day does not just feel worse. It produces worse work. Neuroscience research on task switching consistently shows that refocusing after an interruption takes fifteen to twenty minutes. If you are interrupted more frequently than that — and most knowledge workers are — you never reach the sustained attention that hard problems require. You are paying a recovery tax on every ping, and the tax compounds.

The batched day does not sacrifice responsiveness. Your messages get answered, your colleagues get unblocked. The difference is that you have organized the cost. Instead of paying it unpredictably throughout the day, you pay it deliberately in defined windows.

Inbox zero as a practice, not a score

Inbox zero is frequently misunderstood as an aesthetic — keeping the count at zero so you can feel tidy. That reading turns it into a compulsive habit (check constantly, archive aggressively, chase the number) rather than a discipline.

The original idea, from Merlin Mann who coined the phrase, is about decisions. Every item in your inbox is a deferred decision. If you read a message and leave it there without acting, you will read it again. And again. Each re-read is wasted time — you are paying the cost of processing without taking any action.

Inbox zero as a practice means: touch each message once, during a message window, and decide immediately. The decision options are simple: reply now, flag it for later action with a note, delegate it, or delete it. What you do not do is read it, feel vaguely anxious about it, and leave it sitting there to haunt your peripheral vision.

This requires your message windows to be actual working sessions. Open the inbox with the intention of clearing it. Block the time. Do not browse it passively between other tasks.

A useful heuristic: if a response takes less than two minutes, write it immediately. If it requires real thought — research, a decision you need to think through, input from someone else — flag it and handle it in a separate work block, not in the middle of clearing your inbox. Conflating “reading messages” with “doing the work those messages require” is what makes message windows balloon into hours.

The confidence to be unreachable

The deeper obstacle is not technical. You can turn off Slack notifications in about forty seconds. The real obstacle is the belief — sometimes explicit, often just ambient anxiety — that being unreachable signals disengagement. That a long response time means you are not a team player. That if something important happens and you are not there to react, you will have failed.

This belief is worth examining carefully, because it is often not grounded in anything your organization has actually said. It is a norm that emerged from the early adoption of always-on communication tools, and most organizations never explicitly decided it was good policy. They just defaulted to it.

If you are a few years into your career, you probably cannot unilaterally restructure how your whole team communicates. But you can make a smaller change: be explicit about your availability windows, respond thoroughly when you are present, and see whether the world ends. It usually does not. Colleagues adapt. Managers notice the quality of your work more than the speed of your Slack replies.

If you are in a position to set norms for others — a team lead, a manager, anyone who influences how people collaborate — the calculus is sharper. The norms your team operates under were set by someone. If they were set by default, they can be reset by intention. The question is whether you trust your team enough to let them work.

Notifications are not a neutral feature of modern work. They are a design choice made by platforms that benefit from engagement, adopted by organizations that never thought carefully about the cost. The cost is paid by you, in fragments of attention, across every working day.

You can opt out. The only thing stopping you is the assumption that you cannot.