Triage: Crisis Planning in Action.
How to prioritise when everything feels urgent or heading into crunch
There are periods in planning where the gap between what needs doing and what can actually be done becomes painfully obvious. Deadlines are closing in, tickets are piling up, stakeholders all want certainty, and the team is either already close to burnout or plainly heading there. At that point, the problem is no longer whether the work matters. The problem is that there is too much of it, not enough time, and no sensible way to give everything equal attention.
This is what crunch looks like. Sometimes it is self-inflicted. Sometimes it is the result of poor planning. Sometimes it was visible from a long way off and still could not be avoided. And sometimes it arrives because circumstances changed faster than the plan could. Whatever the cause, once you are in crunch, the same reality applies: everything is not solvable.
That is the point where many teams make the wrong move. They continue to behave as though everything can still be delivered, provided everybody works slightly harder, stretches slightly further, and absorbs slightly more pressure. In practice, this usually means decisions are avoided rather than made. The team stays busy, but the work becomes reactive, people get pulled in too many directions, and eventually the sacrifices happen anyway, only this time without any real control over them.
Crunch is a little like a natural disaster. You may have seen it coming. You may even have prepared for it. But once it arrives, the job changes. The question is no longer how to do everything. The question is how to allocate limited resources in a way that protects the most important outcomes.
Fortunately, there is already a concept for this which we can borrow from the emergency services: Triage.
I like this as a name for a crisis planning session, because it’s the right level of grandiose for businesses - everyone loves a snappy name for something!
Now, obviously, project planning is not emergency medicine. No one is going to die because a feature slips, a report is delayed, or a piece of internal tooling does not make the release. The analogy is not about the severity of the consequences. It is about the nature of the decision.
In both cases, demand exceeds capacity. In both cases, there is not enough time or resource to deal with everything at once. In both cases, some things have to be addressed immediately, some can wait, some can be handled quickly, and some will need to be sacrificed altogether.
That is why triage is a useful model for planning under pressure. It gives you a structured way of deciding what gets attention now, what gets managed next, what can be resolved with minimal effort, and what should be consciously de-prioritised. More importantly, it forces honesty. It forces you to admit that when resources are constrained, not everything can be treated as equally deserving.
If everything is important, then nothing truly is.
Before we go into how we manage tasks, I ask you to remember the main rule of triage:
You can’t save everything. Put your effort where it is going to make the biggest impact and where it is truly needed.
The four-tag model
The simplest way to think about triage in planning is through four tags: Green, Yellow, Red, and Black. The value of the model is not in its sophistication. It is in the fact that it gives teams a common language for making difficult allocation decisions quickly.
A Green Tag is for work that is easy to fix, low risk, and relatively cheap to complete. These are the tasks where the path is clear and the effort is small. A minor bug, a straightforward stakeholder response, a small process correction, or a tidy piece of work that removes friction without demanding much coordination all fit here. Green work is worth doing because it can reduce noise and clear clutter, but it should not be allowed to dominate simply because it feels satisfying. One of the easiest mistakes in crunch is to over-index on easy wins while harder, more consequential work is left untouched.
A Yellow Tag is for work that is difficult but manageable. It matters, and it does need to happen, but it is not yet critical. Yellow items often involve planning, ownership, and judgement rather than immediate intervention. This is where a lot of sensible delivery work sits: process improvements, technical debt with known consequences, work that supports the deadline but is not currently threatening it, and issues that will become serious if ignored for too long. Yellow is often the most important category to manage well, because neglected Yellow work is what tends to turn into Red work later.
A Red Tag is for work that is critical. These are the tasks or issues where the cost of delay is too high. A major blocker to delivery, a serious operational failure, a compliance risk, a customer commitment that cannot slip, or a key dependency that has failed would all belong here. Red work is not simply “important”. Everything in a crunch period is important to somebody. Red work is the work where waiting makes the situation materially worse. These are the things that must be dealt with first, not because they are the loudest, but because they are the most dangerous to ignore.
A Black Tag is for work that is de-prioritised. This is usually the most uncomfortable category, because it forces teams to be explicit about sacrifice. Black-tagging something does not mean it has no value. It means that, under the current constraints, it cannot justify the time, effort, or attention it would require. A nice-to-have feature, a piece of polish, an interesting but non-essential improvement, or a request that matters in theory but not enough in practice may all belong here. Black is what makes triage real, because without a category for deliberate sacrifice, most teams simply continue pretending that everything is still alive.
Triage in action
The most useful way to run triage is to get the key decision-makers in a room with the people closest to the delivery work. That usually means the relevant stakeholders, the person accountable for the deadline, and enough of the delivery team to understand the actual effort and dependencies involved. The point is to bring together authority, context, and reality. If any one of those is missing, the exercise tends to become either political or fictional.
Close the laptops, hide the ticket boards, and stop looking at whatever system has been tracking the work so far. That sounds performative, but it matters. The purpose of triage is not to be led around by tooling. It is to make a fresh judgement about what genuinely needs to happen between now and the deadline. Existing boards often carry too much baggage: old assumptions, stale priorities, and items that are still present only because no one has explicitly killed them.
Write down every item that the room believes must happen before the deadline, along with an estimated amount of time or effort for each. This should be done from collective memory and understanding first, not from the backlog. There is a useful discipline in that. If nobody in the room can bring an item to mind, it is at least worth questioning whether it really belongs in a crunch conversation. That does not mean memory is perfect, but it does force attention onto the work that people actually believe is central.
Review the list once, clean it up, remove duplicates, and challenge vague wording. By this point, the room should be producing a list of concrete things rather than categories of aspiration. “Improve reporting” is too loose. “Complete weekly finance export automation” is something a team can actually triage.
Lock the list. This now becomes the working set for the deadline. This is important because triage falls apart if new items are constantly being smuggled in halfway through the discussion. Once the list is locked, the default assumption should be that this is the full field of work being considered.
Everyone in the room to assigns a tag to each item: Green, Yellow, Red, or Black. This should be done individually first, before the discussion starts to converge opinions. The point is to get an honest first pass rather than a socially negotiated one.
Use the most common tag as the starting position for each item. Where there is a tie, it is usually better to default to the middle rather than jump immediately to the extreme. In practice that means leaning towards Yellow unless there is a compelling reason not to. There is one important rule here: something is not truly Red unless the room broadly agrees that it is. If half the room sees an item as ordinary work and the other half sees it as a crisis, then the first thing missing is not effort but shared understanding.
Discuss the items where there is disagreement. Especially the ones on the boundary between Yellow and Red, and the ones that some people want to Black-tag but others are still trying to save. This is where the value of the room shows up. The discussion is not there to defend territory. It is there to expose assumptions, clarify consequences, and force the trade-offs into the open.
Review the list against both the tags and the available time. This is the moment of truth. If the Reds and Yellows together already exceed the capacity left before the deadline, then the team is still pretending. More has to move to Black, more scope has to be cut, or the deadline itself has to be treated as the thing that is unrealistic.
Finally, prioritise order of action. We have the tags, we know why each item is important against our goal. The final step is to order each item in terms of both importance and our order of operations.
There is also a useful diagnostic here. If you run the exercise and end up with no Black tags at all, then one of two things is probably true:
you do not actually need triage
you are doing it badly.
In a genuine crunch situation, there should be work that does not survive. If everything remains alive, then the team has not made any real sacrifices and has simply re-labelled the problem.
If this happens you aren’t done and need to force honesty. How I solve this is to get each person in the room to choose two items they would Black tag, then discuss which of these are going to be black tagged. The discussion that follows is usually far more revealing than the first pass, because it forces people to confront what they are really willing to let go.
You can’t save everything, something needs to be black tagged. My rule of thumb is if 20% of your items are black tagged - you have a deliverable plan. For example in the 100 hour business I had to do triage recently, for the 17 items I mapped out this was my tag allocation
Green - 1 Tag
Yellow - 5 Tags (First Pass) - 4 Final Tags
Red - 6 Tags (First Pass) - 5 Final Tags
Black - 5 Tags (First Pass) - 7 Final Tags
On review 1 red tag moved to yellow, and two yellows moved to black – I ended up with 40% of my items as black tags.
I have found that you know you will have triaged correctly if honestly, you don’t like it. I am attached to the black tags, I want to do the black tags - but I don’t have time for the black tags when there are more critically important functionalities I need to build.
How to think during triage
Triage only works if the thinking is disciplined. The aim is not to defend your own area, rescue your favourite item, or make sure nothing with your name on it gets dropped. The aim is to make the deadline survivable by being honest about consequence, effort, and capacity.
You need to ask yourself hard questions, and get your thinking to align with the objective of this exercise: creating a manageable plan given the deadline.
So for each item ask:
What happens if it does not get done? This is the question that cuts through theatre very quickly. Some things genuinely threaten delivery, trust, stability, or business outcomes. Others are simply desirable. During triage, those two categories cannot be treated as equivalent.
Is this item genuinely urgent or merely attached to somebody anxious, senior, or loud? These are not the same thing. A great deal of work acquires the appearance of urgency because of stakeholder pressure, but triage requires a more disciplined standard. If delay does not materially worsen the outcome, it should be difficult to justify calling it urgent.
How much effort does the item actually represent? In crunch, feasibility matters as much as value. A task may well be worthwhile, but if it is large, uncertain, and likely to consume disproportionate capacity, that should influence how it is tagged. Equally, smaller items with clear benefit may deserve to be handled quickly, particularly where they remove friction elsewhere.
Does the item meaningfully supports the outcome the team is trying to protect? This is where triage becomes different from general prioritisation. In ordinary planning, there is room to advance several objectives at once. In crunch, there usually is not. The team needs to be clear what it is trying to preserve: a release date, system stability, customer trust, legal compliance, or simply the ability to get to the deadline without breaking the people doing the work. If an item does not support that outcome, it should struggle to survive triage.
That is also why triage has to be collective rather than personal. Individuals are usually too close to their own work to assess it cleanly. Triage works best when the room is capable of judging items not by ownership or politics, but by consequence. That requires some detachment, some honesty, and a willingness to accept that good ideas can still be wrong for the moment.
The hard truth about triage
The real value of triage is not that it makes difficult periods pleasant. It does not. The work is still constrained, the deadline is still real, and compromise is still unavoidable. What triage does is replace chaos with deliberate judgement.
That matters because the alternative is rarely neutral. If you do not consciously choose what to sacrifice, sacrifice still happens. It just happens badly. Quality gets cut in the wrong places. Deadlines slip unpredictably. Teams burn out while working on the wrong things. Trust erodes because expectations were never reset honestly enough. Avoiding the decision does not remove the trade-off. It simply makes the trade-off uncontrolled.
This is why triage matters. It gives you a way to look at a crowded field of work and decide, with some discipline, what gets saved, what gets scheduled, what gets cleared, and what gets left behind. That is not comfortable, but it is what leadership looks like when resources are constrained and the cost of pretending has become too high.
Triage is really about honesty
At a surface level, triage looks like a method for handling urgency. In practice, it is really a method for forcing honesty. It forces honesty about the deadline, honesty about the work, honesty about the team’s capacity, and honesty about the fact that some things will not make it through intact. Most planning does not go wrong because people do not care. It goes wrong because they try to preserve too much for too long.
Good planning is not about convincing yourself that everything can still be done. It is about recognising when it cannot, and being disciplined enough to decide what survives. That is what triage is for. It is not just a way to order work. It is a way to make necessary sacrifices deliberately, rather than absorbing them accidentally through slippage, burnout, and deteriorating quality. In that sense, triage is not simply a response to pressure. It is what stops pressure from turning into failure.
This gives you a stronger final note because it ends on a broader principle rather than just restating the framework.
Grand. Thanks for your time.