What Good Triage Looks Like

There is a moment in every emergency room when someone has to decide who gets seen first. The guy clutching his chest goes ahead of the woman with the sprained ankle, who goes ahead of the teenager with the suspicious rash. Nobody is happy about this. The woman’s ankle genuinely hurts. The teenager is genuinely worried. But the guy clutching his chest might genuinely die, and the system exists to prevent that outcome even when it means everyone else waits.

This is triage. Not prioritization — triage. The distinction matters. Prioritization is about ordering a list from most important to least important. Triage is about making rapid, sometimes brutal decisions about resource allocation when demand exceeds capacity. Prioritization assumes you will eventually get to everything. Triage assumes you will not.

If you have been following along in this book, you already know where this is going. Your information environment is an emergency room at capacity. The demand — articles, papers, emails, newsletters, Slack messages, podcast episodes, reports, social media threads, notifications — exceeds your capacity to process it. Every single day. And it is not getting better.

The good news: triage is a learnable skill. Doctors are not born knowing how to do it. They learn frameworks, practice applying them under pressure, and develop intuition that gets faster and more accurate over time. You can do the same thing with information. The better news: unlike emergency medicine, getting information triage wrong rarely kills anyone.

The Three Buckets

Medical triage traditionally uses a simple categorization system. The specific implementations vary, but the core logic is always some version of three buckets:

Immediate: Requires attention right now. Delay creates serious consequences.

Delayed: Requires attention eventually. Can wait without significant harm.

Expectant: Will resolve on its own, or the investment required exceeds the likely return. (In emergency medicine, this category has a grimmer interpretation. In information management, it just means “ignore this.”)

That is it. Three buckets. Not seventeen priority levels, not a color-coded matrix with weighted scoring criteria, not a Kanban board with twelve columns. Three buckets. The power of triage lies in its simplicity, and if you overcomplicate it, you end up spending more time triaging than processing.

Let me translate these into information terms:

Process Now: Information that is directly relevant to something you are actively working on, time-sensitive in a way that actually matters, or requires a decision within the next 24-48 hours. This is your chest-pain category.

Defer: Information that is relevant to your work or interests but does not require immediate attention. The new research paper in your field that you want to read carefully. The industry analysis that will be useful next quarter. The long-form article a colleague recommended. This is your sprained-ankle category — it needs attention, but it can wait for the right time.

Ignore: Everything else. And I mean everything else. The thread that is interesting but tangential. The newsletter you subscribed to two years ago and have not opened in months. The article with a provocative headline about a topic you do not actually need to understand. The suspicious rash. It will probably be fine.

If you are doing triage correctly, the Ignore bucket should be the largest by a significant margin. If it is not, you are either working in an unusually information-dense role or you are not being honest with yourself about what actually matters.

Urgency Is Not Importance

One of the most consistently useful mental models I have encountered in productivity thinking — and I have encountered a depressing number of them — is the Eisenhower Matrix. You have probably seen it: a two-by-two grid with urgency on one axis and importance on the other. It is attributed to Dwight Eisenhower, though like most things attributed to famous people, the attribution is somewhat dubious.

The framework’s insight is simple: urgency and importance are different dimensions, and we systematically confuse them. Something can be urgent without being important (the Slack message marked with a red exclamation point about a meeting room change). Something can be important without being urgent (the research paper that would fundamentally change your approach to a project, but which will still exist next week). And we have a deeply wired tendency to respond to urgency at the expense of importance.

For information triage, this maps directly:

Urgent and Important: The security advisory about the tool your team uses in production. The regulatory change that affects your current project. The breaking news about your industry that clients will ask about today. Process now.

Important but Not Urgent: The in-depth analysis of trends in your field. The book your mentor recommended. The technical deep-dive on a technology you will be adopting next quarter. Defer, and schedule time for it.

Urgent but Not Important: The trending topic on social media that everyone is talking about but that has no bearing on your work or decisions. The “BREAKING” news that will be forgotten in 48 hours. The urgent-sounding email that is actually just someone else’s poor planning. Ignore, or at most skim the headline to confirm it does not matter.

Neither Urgent nor Important: I trust you can identify these. They are the ones you spend the most time on.

The challenge is that information sources are specifically designed to make everything feel urgent. “Breaking news” banners. Push notifications. Exclamation points. Red dots. The entire visual language of modern information delivery is calibrated to trigger urgency responses. Your triage system needs to be strong enough to override those signals, which means it needs to be based on something more durable than your in-the-moment feelings about what seems pressing.

Building Your Triage Framework

A triage framework is a set of questions you ask about each piece of incoming information to determine which bucket it belongs in. The questions need to be fast — you should be able to answer them in under 30 seconds for most items — and they need to be specific to you.

Here is a starter framework. Modify it until it fits your actual life:

Question 1: Does this relate to something I am actively working on or responsible for?

If no, it is almost certainly Defer or Ignore. The key word is “actively.” Not “might work on someday.” Not “tangentially related to my field.” Actively working on. Right now.

Question 2: Is there a genuine time constraint?

And by genuine, I mean: will something bad actually happen if I do not process this information today? Not “will I feel behind” or “will I miss the conversation.” Will an actual consequence occur? If you have to think hard about what the consequence would be, the answer is no.

Question 3: Will this information become significantly less useful if I delay processing it?

Some information is perishable. Market data loses value quickly. News about a developing situation changes by the hour. A vulnerability announcement needs action before exploitation. Other information is durable. A research paper published last week will be equally useful next week. A book recommendation does not expire. Understanding which category you are looking at prevents false urgency.

Question 4: Am I the right person to process this, or can it be delegated or shared?

This one is often overlooked. If someone sent you an article because they thought it was interesting, that does not obligate you to read it. If an email contains information that someone on your team is better positioned to evaluate, forward it. Triage includes routing, not just categorizing.

Question 5: What is the worst realistic outcome if I ignore this entirely?

Not the worst imaginable outcome. The worst realistic one. If the answer is “nothing much” — and it usually is — you have your answer.

Run through these five questions, and most information will sort itself fairly quickly. The items that survive all five questions and still land in Process Now are the ones that actually deserve your immediate attention.

Speed-Reading vs. Deep-Reading

There is a persistent myth that speed-reading is the answer to information overload. If you could just read faster, you could get through everything. This is like saying that if you could just eat faster, you could eat at every restaurant in the city. The problem is not speed. The problem is selection.

That said, there is a legitimate distinction between different reading modes, and part of good triage is matching the right mode to the right material.

Scanning (5-15 seconds per item): You are looking at headlines, subject lines, abstracts, and first sentences. The goal is to determine whether this item deserves further attention at all. This is not reading — this is sorting. You should be able to scan 50-100 items in ten minutes.

Skimming (1-3 minutes per item): You are reading the introduction, the conclusion, the section headings, and any text that is bold, highlighted, or otherwise marked as important. You are getting the shape of the argument without every detail. This is appropriate for items in your Defer bucket that you want to understand well enough to decide whether they warrant full reading later.

Focused Reading (10-30 minutes per item): You are reading the whole thing, but you are not taking notes or deeply engaging. You are absorbing the content at a normal reading pace. This is appropriate for most items in your Process Now bucket.

Deep Reading (30+ minutes per item): You are reading carefully, taking notes, cross-referencing with other sources, thinking critically about the arguments. This is appropriate for a small number of items that are genuinely important to your work or thinking. If you are doing deep reading on more than a few items per week, you are either in an unusually research-intensive role or you are being insufficiently ruthless in your triage.

The critical insight: these are not four points on a spectrum of “how much effort.” They are four different activities with four different purposes. You should consciously decide which mode you are in before you start, not drift from scanning into deep reading because something caught your attention. Unplanned mode-switching is how two hours disappear.

The Two-Minute Rule, Adapted

David Allen’s Getting Things Done methodology includes the “two-minute rule”: if a task takes less than two minutes to complete, do it immediately rather than adding it to your task list. The overhead of tracking and returning to the task exceeds the time to just do it.

This adapts beautifully for information triage:

If you can fully assess and process a piece of information in two minutes or less, do it now during triage.

This means: if you can read a short article, extract the one relevant fact, and file or discard it — do it. If you can read an email, determine it requires no action, and archive it — do it. If you can look at a notification, confirm it is irrelevant, and dismiss it — do it.

What this does NOT mean: starting to read a long article because “I’ll just glance at it.” You know how that ends. You are eight paragraphs in, you have been reading for twelve minutes, and you have abandoned your triage session for an unplanned deep-read. The two-minute rule is a boundary, not a suggestion.

In practice, this means that during a triage session, you will fully process many small items and sort the rest into Defer or Ignore. Your triage session becomes both a sorting activity and a processing activity for quick items. This is efficient. It also provides the satisfying feeling of “getting things done” that makes triage sessions feel productive rather than purely administrative.

Preview, Skim, Full-Read: A Processing Pipeline

Let me formalize this into a pipeline, because pipelines are easier to follow than principles.

Stage 1: Preview

You look at the metadata only. Title, source, author, date, length. For emails, the subject line and sender. For articles, the headline and publication. For papers, the title and abstract.

At this stage, you are making one decision: does this advance to Stage 2, or does it go into Ignore?

Criteria for advancing:

• The source has established credibility for this topic
• The title suggests relevance to your active priorities
• The author is someone whose work you have found valuable before
• The topic matches something in your defined information priorities (more on this in Chapter 12)

Everything else stops here. Deleted, archived, dismissed. Gone.

Stage 2: Skim

You spend 60-90 seconds getting the shape of the content. For articles, read the first paragraph, the last paragraph, and the subheadings. For papers, read the abstract and conclusion. For emails longer than a few sentences, read the first and last paragraphs and scan for action items.

At this stage, you are making two decisions: (1) does this go into Process Now or Defer? and (2) what reading mode does it require?

If it is Process Now, you either handle it immediately (if it passes the two-minute rule) or add it to today’s reading queue with the appropriate mode noted.

If it is Defer, you file it somewhere retrievable — a read-later app, a bookmark folder, a note in your task manager — with enough context that future-you will know why past-you saved it.

Stage 3: Full-Read

This happens outside of triage, during dedicated reading time. You are processing items from your Process Now queue and selectively pulling from your Defer archive. The reading mode (Focused or Deep) was determined during Stage 2.

This pipeline is fast. With practice, you can Preview 100 items, Skim 20 of them, and identify 5-8 for Full-Read in about 30 minutes. That is a reasonable daily triage session.

The Triage Checklist

Because checklists work better than principles when you are building a habit, here is a concrete one. Print it out if that helps. Tape it to your monitor. Whatever makes you actually use it.

Before starting triage:

• Close everything except your information sources and your filing system
• Set a timer (30 minutes for daily triage, 60 minutes for weekly review)
• Have your “Process Now” queue, “Defer” archive, and “Ignore/Delete” action ready

For each item in your inbox/feed/queue:

• Preview: Read only the metadata (title, source, author, date)
• Ask: Does this relate to my active priorities?
• If no → Ignore. Archive or delete. Move on.
• If yes → Skim: Spend 60-90 seconds on structure and key points
• Ask: Is this time-sensitive? Do I need it today?
• If yes and it passes the two-minute rule → Process it now
• If yes but it needs more time → Add to today’s reading queue
• If no → File in Defer archive with a note about why it matters

After triage:

• Review your Process Now queue — is it realistic for today?
• If too many items, re-triage the queue using the urgency/importance framework
• Schedule reading blocks for the items that require Focused or Deep reading
• Note anything from Defer that should be scheduled for this week

Weekly (in addition to daily):

• Review the Defer archive — re-triage everything older than two weeks
• Items that have sat in Defer for two weeks without being read probably belong in Ignore
• Adjust your Preview criteria based on what you actually found useful this week
• Identify any patterns: are certain sources consistently landing in Ignore? Unsubscribe.

When Triage Happens

The single most destructive pattern in information management is continuous triage. Checking email constantly. Glancing at news throughout the day. Keeping Slack visible at all times. Monitoring social media in the background. This is not triage. This is surveillance, and it is exhausting.

Good triage happens at fixed times. For most people, two to three triage sessions per day is sufficient:

Morning triage (15-20 minutes): Process overnight accumulation. Email, messages, news. Sort everything into the three buckets. Handle quick items. Build your reading queue for the day.

Midday triage (10-15 minutes): Quick check for anything time-sensitive that arrived since morning. This is a rapid scan, not a full processing session. If nothing urgent has arrived, skip it entirely.

End-of-day triage (10-15 minutes): Process afternoon accumulation. Handle quick items. Move anything remaining to tomorrow’s morning triage. Clear your inboxes to reduce overnight anxiety.

Between these sessions, your information channels should be closed. Not minimized — closed. Not muted — closed. The entire point of fixed triage times is that you are not triaging the rest of the time. You are doing actual work, which is the thing that information is supposed to support.

“But what if something urgent comes in and I miss it?”

If something is genuinely urgent — server down, building on fire, you know the kind of thing I mean — it will reach you through a channel you cannot ignore. A phone call. Someone walking to your desk. A pager. If the only way you would find out about an urgent situation is by constantly monitoring your email, then either the situation is not as urgent as you think, or your team needs to set up better alerting.

The fear of missing something urgent is almost always disproportionate to the actual frequency of urgent things. Most people, in most roles, encounter genuinely urgent information a few times per month. The daily drip of pseudo-urgent notifications trains you to expect urgency constantly, which keeps you in a state of continuous low-grade vigilance that is terrible for both your cognitive function and your well-being. Fixed triage times break that cycle.

A Sample Daily Workflow

Let me walk through what this actually looks like in practice. This is based on my own workflow, adapted for a generic knowledge worker. Your specific tools and sources will differ, but the structure should translate.

6:30 AM — Morning Triage

I open three things: email, my RSS reader, and my team’s messaging tool. Nothing else. No social media, no news sites, no “just a quick check” on anything.

Email (10 minutes): I scan subject lines and senders. Anything from my direct team or active project stakeholders gets opened. Everything else gets a Preview-level assessment based on the subject line.

• Quick replies (two-minute rule): done immediately
• Items requiring longer responses: starred and added to my task list
• FYI emails and newsletters: Preview → Skim if relevant → Defer or Ignore
• Automated notifications: scanned for anomalies, then archived

RSS reader (5 minutes): I subscribe to about 30 feeds, which typically accumulate 100-150 new items overnight. I scan headlines only. Items that pass the Preview filter get saved to my read-later app. I mark everything as read. This is the part where ruthlessness matters most — if I save more than 10 items, I am being too generous.

Team messages (5 minutes): I read all direct messages and mentions. I scan channel activity for anything relevant to my current work. I respond to quick items. I note anything that needs a longer response.

Total morning triage: 20 minutes. I now know what today’s information landscape looks like and have a queue of 5-10 items for reading time.

9:00 AM — First Reading Block

I block 45 minutes for reading from my Process Now queue. This is not triage — this is focused processing. I work through the items I identified during morning triage, in order of importance. If I finish early, I pull one item from Defer. If I do not finish, remaining items stay in the queue for tomorrow (they clearly were not urgent enough to justify more time today).

12:30 PM — Midday Scan

Five minutes. I glance at email for anything marked urgent or from key stakeholders. I check team messages for direct communications. If nothing urgent has appeared, I close everything and move on.

4:30 PM — Afternoon Triage

Similar to morning triage but shorter (10-15 minutes). I process afternoon email, check for anything that needs a response before end of day, and clear my queues. Items that arrived during the day and were not time-sensitive go into tomorrow’s morning triage.

Total daily information processing: approximately 80-90 minutes

This includes triage AND reading. For most knowledge workers, this is sufficient to stay informed and responsive. If you are currently spending more than two hours per day on information processing (and if you track it honestly, you probably are), the difference is time recovered for actual work.

A Sample Weekly Workflow

In addition to the daily rhythm, I do a weekly information review. This takes about an hour and happens on Friday afternoons, when I am least likely to be interrupted and most likely to be in a reflective mode.

Friday 3:00 PM — Weekly Review

Defer archive review (20 minutes): I look at everything I saved to my read-later app during the week. Some of it now feels less relevant than it did when I saved it — that is fine, delete it. Some of it is genuinely worth reading — I schedule time for it next week. Anything that has been in the archive for more than two weeks and I still have not read it gets deleted. If it was important, it will come back.

Source audit (10 minutes): I look at which information sources consistently provided useful content this week and which consistently did not. If a newsletter has been Ignored three weeks running, I unsubscribe. If an RSS feed has not produced a single item I wanted to read in a month, I remove it. This is how you keep your information inputs manageable over time — regular pruning.

Priority check (15 minutes): I review my active projects and responsibilities. Have my information priorities shifted? Is there a new topic I need to be tracking? An old one that has resolved? I adjust my triage criteria accordingly. This prevents the slow drift where your information intake stays the same while your actual needs change.

Deep reading selection (15 minutes): From my Defer archive and any recommendations I have received, I select 2-3 items for deep reading next week. I schedule specific time blocks for them. This ensures that important-but-not-urgent reading actually happens instead of perpetually sitting in a “someday” queue.

When the System Breaks Down

It will break down. You will have a week where three deadlines converge and triage goes out the window. You will have a day where a genuinely urgent situation requires continuous monitoring. You will have a morning where you open your RSS reader with the best intentions and emerge two hours later from a Wikipedia rabbit hole about the history of pneumatic tube mail systems.

This is normal. The point of having a system is not that you follow it perfectly every day. The point is that you have a default to return to. When the chaos subsides, you do not have to reinvent your information management approach from scratch. You just go back to the system.

A few guidelines for recovery:

After a disrupted day: Do not try to catch up on everything you missed. Do a single extended triage session (45-60 minutes) to sort the backlog, then resume normal operations. Most of what accumulated during the disruption has either been handled by someone else, resolved itself, or is no longer relevant.

After a disrupted week: Declare information bankruptcy on everything non-critical. Mark all feeds as read. Archive all non-urgent emails. Start fresh on Monday with normal triage. If something important was in that pile, it will resurface.

After falling off the system entirely: Start with one component. Just do morning triage for a week. Once that is habitual, add the reading block. Then the weekly review. Trying to implement the full system after falling off it is how you abandon it permanently.

Triage by Content Type

Not all information arrives in the same shape, and different shapes demand different triage approaches. Let me walk through the major content types you are likely to encounter and how to handle each one efficiently.

Email

Email is the most universally dreaded information channel, and for good reason: it mixes high-priority direct communications with low-priority newsletters, automated notifications, and messages that were sent to you only because someone hit “Reply All.” The triage approach for email is aggressive filtering by sender and type before you even look at content.

First pass — Sender filter (30 seconds): Scan the sender column. Anything from your direct manager, active project stakeholders, or direct reports gets opened immediately. Everything else waits for the second pass. This is not about the importance of the content — it is about the relationship. These people need to know you are responsive, and their messages have the highest probability of being genuinely urgent.

Second pass — Type filter (2 minutes): Group the remaining emails mentally: newsletters, automated notifications, FYI/CC emails, and direct messages from people outside your immediate circle. Process automated notifications first (they are usually either immediately actionable or immediately ignorable). Then newsletters (batch for later or delete). Then FYI emails (skim for anything that changes your picture). Then everything else.

Third pass — Content triage (remaining time): For anything that survived the first two passes, apply the standard triage framework. Most items will be two-minute-rule quick. A few will need to be queued for a longer response.

The key insight: do not read emails in chronological order. Read them in priority order, determined by sender and type. Chronological order means you spend your freshest attention on whatever arrived first, which is random.

News and Current Events

The news is specifically designed to feel urgent. Everything is breaking. Everything is developing. Everything demands your attention right now. Almost none of it actually does.

For news triage, the most important question is: “Will this still matter in a week?” If the answer is no, you can safely ignore it during today’s triage and catch up on anything genuinely significant during your weekly review. If the answer is yes, it deserves a skim. If the answer is “this will matter in a month,” it deserves a full read.

In practice, this means you should get your news from sources that aggregate and contextualize rather than sources that report in real time. A morning briefing newsletter is almost always more useful than a live news feed. The live feed gives you the feeling of being informed; the morning briefing actually informs you.

Research Papers and Technical Documents

These are almost never urgent and almost always important. The correct triage action for a research paper is nearly always “Defer” — specifically, defer to a dedicated reading block where you can give it the attention it deserves.

The exception is when a paper directly addresses a question you are actively trying to answer right now. In that case, it goes into Process Now, but even then, the processing should happen during a reading block, not during triage. During triage, you are only deciding that this paper should be read — you are not reading it.

For triage purposes, the abstract and conclusion of a paper tell you almost everything you need to make the triage decision. Read those, determine whether it is relevant to your active work, and sort accordingly. The methodology and results sections are for the reading block.

Social Media and Discussion Threads

These are the highest-volume, lowest-signal information sources most people encounter. The triage approach should be aggressive: give social media a fixed, short time budget (10-15 minutes per day, maximum), and when the time is up, close it regardless of what you have or have not seen.

Do not try to “catch up” on social media. It is a stream, not a queue. You dip in, you see what is there, you dip out. If something important happened on social media, you will hear about it through other channels — if you do not, it was not actually important.

The one exception: if social media is a primary channel for your professional community (common in some fields, including parts of tech and academia), treat it like a focused RSS feed. Follow a curated list of relevant accounts, ignore the algorithmic feed entirely, and process it during a dedicated triage slot.

Slack, Teams, and Workplace Chat

Workplace chat is the most insidious triage challenge because it feels like it requires continuous attention. Someone might message you at any time, and the expectation of rapid response creates pressure to monitor constantly.

The solution is the same as for all other channels: fixed triage times with closed channels in between. Check your direct messages and mentions three times per day. Scan channels relevant to your active work once or twice. Everything else can wait.

If your workplace culture genuinely requires faster response times than this, negotiate it explicitly: “I check messages at 9, 12, and 4. If something is truly urgent, call me.” Most people will never call, which tells you how rarely something is actually urgent.

Triage Metrics: How to Know If It Is Working

Any system benefits from measurement, and triage is no exception. Here are a few metrics worth tracking, at least informally, during your first month:

Time spent triaging per day. Track this honestly. If it is consistently over 60 minutes, your sources are too numerous or your criteria are too loose. If it is under 15 minutes, you might be skipping triage sessions or not being thorough enough.

Items in each bucket. Rough counts are fine. You should be Ignoring 60-80% of incoming items, Deferring 15-25%, and Processing Now 5-15%. If your Process Now bucket is consistently above 20%, you are either defining “now” too broadly or you are in a genuinely unsustainable information environment that needs structural intervention (fewer subscriptions, fewer channels, delegation).

Defer archive size. If your Defer archive grows indefinitely, you are saving more than you are reading. This is not a triage success — it is procrastination with extra steps. A healthy Defer archive turns over regularly: items come in, get read or get deleted, and the total count stays roughly stable.

Missed important items. This is the one you are worried about. Track any instance where something important reached you late or not at all because your triage filtered it out. In practice, this will happen rarely — but when it does, it is an opportunity to adjust your criteria. Was the miss because of a bad source list, an overly aggressive ignore rule, or a genuine edge case?

Reading time quality. Subjective but important. After your reading blocks, ask yourself: was what I read worthwhile? If you consistently feel that your reading time was well spent on content that was relevant and high-quality, your triage is working. If you regularly feel like you read things that were not worth the time, your triage is letting too much through.

Do not obsess over these metrics. The point is to create a feedback loop that helps you improve the system over time, not to add administrative overhead to an already-complex workflow. A quick mental check-in during your weekly review is sufficient.

The Emotional Component

I have been talking about triage as if it is purely a rational process, and I should correct that before we move on. There is an emotional dimension to information triage that is worth acknowledging.

Ignoring information feels bad. It triggers a specific anxiety — what if that was important? What if I miss something? What if everyone else knows about this and I do not? This anxiety is real, and it is one of the main reasons people default to continuous monitoring instead of structured triage. Continuous monitoring does not actually work better, but it soothes the anxiety of potentially missing something by replacing it with the certainty of seeing everything (while comprehending little of it).

The antidote is not to suppress the anxiety. It is to build evidence that the system works. After a few weeks of structured triage, you will notice something: you did not miss anything that mattered. The things that were important reached you. The things that did not reach you turned out to be unimportant. The evidence accumulates, the anxiety diminishes, and triage starts to feel like relief instead of sacrifice.

There is also the FOMO dimension — the fear that you are missing out on interesting, enriching, or entertaining content. This is a different issue from missing important content, and it requires a different response. The answer here is honest: yes, you are missing things. You are missing a lot of things. You will always miss most things. The question is not whether you miss things — it is whether the things you do process are the right things for you. A good triage system makes that trade-off deliberately rather than leaving it to chance and algorithms.

Triage for Teams

If you work on a team, individual triage is necessary but not sufficient. Teams need shared triage practices too, because a significant portion of your information overhead comes from the team’s information environment, not just your personal one.

The Rotating Triage Role

In teams I have worked with, one of the most effective practices is designating a rotating “information triage” role. Each week, one team member is responsible for scanning shared information channels — industry news, competitive intelligence, relevant research, stakeholder communications — and producing a brief summary for the team. Everyone else can reduce their personal triage of those channels during that week.

This works because triage is duplicative: if five people on a team are all scanning the same feeds and newsletters, that is five times the effort for the same information. A rotating role means the work gets done once and everyone benefits.

The format for the team triage summary should be simple. One page, maximum. Organized by the team’s shared priority map. Each item gets a one-line summary and a relevance assessment. The whole team reviews it at the start of the week — five minutes, not a meeting, just a shared document.

Shared Triage Criteria

Teams should have explicit, shared criteria for what information gets escalated versus what gets filed. Without this, you get two failure modes: either everyone escalates everything (“just in case you haven’t seen this”) and everyone drowns in forwarded articles, or nobody escalates anything and critical information gets trapped in one person’s inbox.

A simple shared framework: anything that affects active projects, team processes, or immediate deadlines gets shared in the team channel. Everything else gets included in the weekly triage summary if relevant. Nothing gets forwarded to the whole team with “FYI” and no context. If something is worth sharing, it is worth a one-sentence note about why.

The Team’s Information Budget

Teams, like individuals, have a finite capacity for information processing. And teams are worse at managing that capacity than individuals, because the costs of information overload are distributed. When one person forwards an article to the team, it costs them ten seconds and the team ten minutes (five people times two minutes each). The person forwarding does not feel the distributed cost.

Making this explicit helps. Some teams I have worked with have adopted a rough “information budget” — a shared understanding of how many items the team should be expected to process per week from shared channels. The number matters less than the agreement. Once you have acknowledged that the team’s attention is finite and shared, the incentive to be selective about what you share follows naturally.

This is also where team norms around meeting agendas, status updates, and documentation standards intersect with information triage. Every recurring meeting that could be an email is a triage failure at the team level. Every status update that repeats information available in the project tracker is wasted attention. Good team triage is not just about external information — it is about respecting the team’s internal bandwidth too.

Common Triage Mistakes

I have watched enough people try to implement triage systems to know where they usually fail. Here are the patterns I see most often:

Mistake 1: Triaging during reading time. Triage and reading are different activities. When you sit down for a reading block and start by scanning new items to decide what to read, you are doing triage when you should be reading. Triage produces the queue; reading blocks consume it. If you mix them, you end up with a triage session that takes an hour and a reading block that never happens.

Mistake 2: Too many categories. Three buckets. Not five, not seven, not a priority matrix with color coding. Every additional category adds decision overhead to every item you triage. The people who spend twenty minutes designing a seventeen-label tagging system and then spend the rest of the week not using it — I have been that person, and it does not work.

Mistake 3: Treating Defer as a guilt pile. The Defer bucket is not where information goes to make you feel bad. It is a holding area for things that are genuinely worth reading at a scheduled time. If looking at your Defer archive fills you with anxiety, the archive is too large and needs aggressive pruning. If an item has been in Defer for more than two weeks, it has voted with its feet — you are not going to read it, and keeping it around just adds to the psychic weight.

Mistake 4: Not actually closing your channels. “I’ll just keep Slack open in case something urgent comes in” is the death of structured triage. The notification badge on a minimized window is a leash. Close means close: quit the application, close the browser tab, disable the notification. If this feels extreme, try it for three days and see if anything bad happens. It will not.

Mistake 5: Trying to reach inbox zero during triage. Inbox zero is a processing goal, not a triage goal. Triage is about sorting, not completing. If you try to fully process every item during triage, your triage session becomes your entire morning. Sort into buckets, handle the quick items (two-minute rule), and move on. The queue is for later.

Mistake 6: Not adjusting for context. Your triage criteria should flex based on what is happening in your work. During a product launch, competitive intelligence moves to Tier 1. During a strategic planning cycle, industry analysis becomes more urgent. During a normal week, these might be Tier 2 items. Static criteria in a dynamic environment lead to misallocated attention. Review your triage criteria when your work context changes.

Mistake 7: Perfectionism about the system itself. Some people spend so long designing the perfect triage system that they never actually start triaging. The best system is the one you use consistently, not the one that is theoretically optimal. Start with the three buckets, the five triage questions, and fixed triage times. Iterate from there. A crude system used daily beats a sophisticated system used never.

Mistake 8: Comparing your triage to someone else’s. Your colleague who seems to read everything and stay on top of every conversation is either spending more time on information processing than you realize, processing less deeply than you assume, or working in a different role with different demands. Your triage system should be calibrated to your priorities, your role, and your cognitive style. Someone else’s system is informative as a reference, not prescriptive as a standard.

What Good Triage Feels Like

I want to end this chapter with something less structural and more experiential, because I think it matters.

Good triage feels calm. Not complacent — calm. You know what you are processing, you know what you are deferring, and you know what you are ignoring. The decisions are explicit. The uncertainty is bounded. You are not scanning the horizon anxiously for threats; you are checking at defined intervals and then returning to work with confidence that you have not missed anything critical.

Good triage feels fast. The actual triage sessions should feel brisk, even a little ruthless. You are making snap decisions based on clear criteria, not agonizing over each item. If a triage session feels slow and deliberate, your criteria are probably too complex or you are doing too much reading during triage instead of saving it for reading time.

Good triage feels incomplete, and that is fine. You will end every triage session with items in your Defer bucket and a vague awareness that you Ignored things that might have been interesting. This is the correct feeling. Complete coverage is not the goal. The goal is appropriate coverage — the right information, at the right depth, at the right time.

Good triage feels like a skill that gets better with practice. The first week you try it, everything will feel slow and awkward. You will second-guess your Ignore decisions. You will check your email outside of triage times. You will spend too long skimming things that should have been deleted at Preview. This is normal. Like any skill, triage develops through repetition. By the end of the first month, the decisions that took you thirty seconds will take you three. By the end of the third month, triage will feel like a natural part of your day rather than an imposed discipline.

And good triage feels liberating, in a way that is hard to appreciate until you have experienced it. The constant low-grade anxiety of “I should be checking something” is replaced by the calm confidence of “I will check at my next triage time, and that is sufficient.” Your reading time is spent on content that was selected through explicit criteria, not whatever happened to be at the top of the feed. Your attention is allocated by you, not by an algorithm. That is not a small thing.

In the next chapter, we will look at how to use LLMs to make every stage of this pipeline faster and more effective. Because the triage framework I have described here was designed for humans processing information manually, and we now have tools that can dramatically accelerate the Preview and Skim stages in particular. But the framework comes first, because an AI that helps you process information faster is only useful if you know which information is worth processing.