The Triage Lens
How to Prioritise What We Do When We Can’t Do Everything
Most of us operate under a persistent delusion that we can do it all. We look at a performance problem, or a new initiative, or a capability gap, and we draw up a plan that covers every conceivable angle, every nice-to-have element, every supplementary resource and follow-up activity that might, in an ideal world, contribute to a better outcome. Then we run out of time, or budget, or both, and the whole thing collapses into a compromised version of itself where nothing gets done particularly well because we tried to do too much.
To combat this in my own practice, I use a three-tier prioritisation lens that forces me to distinguish between what must happen, what should happen if we can make it work, and what would be nice if we had unlimited time and resources but can be safely deprioritised without undermining the outcome. It’s not a new idea, it borrows from triage medicine, from the MoSCoW method in software development (Clegg, 1994), and from the broader principle that a small proportion of our inputs tend to generate a disproportionate amount of our outputs (Juran, 1951).
The three tiers
The model works across three distinct lenses depending on the nature of the work: what people need to know, what people need to be able to do, and what needs to happen to solve the problem. Each of these lenses breaks down into the same three tiers, and the logic is consistent across all of them.
Tier 1: Must. This is the bare minimum that has to happen for the intervention, the programme, or the solution to succeed. If this tier doesn’t get delivered, the whole effort fails. In knowledge terms, this is what someone absolutely needs to know to perform their role safely, legally, or to the standard required. In skills terms, this is what someone must be able to do on day one, in the field, without support. In problem-solving terms, this is the set of actions that are essential to resolving the root cause.
Tier 2: Should. This tier captures the things that will have a measurable positive impact but are not essential for the minimum viable outcome. In knowledge terms, this is information that deepens understanding, that adds context, that helps someone make better decisions over time. In skills terms, these are the capabilities that improve efficiency, quality, or consistency beyond the baseline. In problem terms, these are the actions that address contributing factors and secondary causes.
Tier 3: Nice to have. This is everything else. It might have some impact, but that impact will be small, inconsistent, and difficult to attribute. This tier is the material that L&D teams love to include because it feels thorough, because it’s interesting, or because someone important asked for it, but it doesn’t meaningfully move the needle. In knowledge terms, it’s background context that enriches but doesn’t change behaviour. In skills terms, it’s the advanced or edge-case capability that most people will rarely use. In problem terms, it’s the peripheral activity that makes us feel comprehensive but contributes little to the resolution.
The critical discipline is protecting tier 1 at all costs. Everything else is negotiable.
Why this matters
Thomas Gilbert (1978), made a distinction that’s central to this idea: the difference between worthy performance and costly behaviour. Gilbert argued that competence should be measured by the ratio of valuable accomplishments to the cost of the behaviour required to produce them. An intervention that tries to cover everything incurs enormous cost, both in design effort and in the load placed on the people going through it, without a proportional increase in the value of what gets accomplished.
Mager and Pipe (1997) reinforced this from a different angle. Their performance analysis model is built around a series of questions designed to prevent organisations from investing in solutions that don’t match the problem. The very first meaningful question in their flowchart is whether the performance gap is worth fixing at all, and one of the recurring themes throughout is the preference for simpler, cheaper interventions over complex, expensive ones. The logic is the same: focus your resources where they will have the most impact, and resist the urge to address everything simultaneously.
The Pareto principle, originally an observation about wealth distribution (Pareto, 1896) and later generalised by Juran (1951) into a tool for quality management, suggests that roughly eighty percent of effects tend to come from roughly twenty percent of causes. The exact ratio varies, but the underlying insight holds across an extraordinary range of contexts: a small number of factors tend to account for the majority of the outcome. In training terms, a small number of knowledge elements and skills tend to account for the majority of performance improvement, and identifying those elements is far more valuable than building a comprehensive programme that covers everything with equal emphasis.
If you’d like to explore how you can use research from psychology, neuroscience, the cognitive sciences and behavioural sciences in your work in L&D, HR, performance enablement or leadership, consider attending this year’s Evidence Informed Practice Conference.
As a reader of this Substack, you can get 25% off your ticket using code CPDW25 at checkout.
Applying the lens
The three-tier lens works best when applied early, during the scoping and design phase rather than as a post-hoc rationalisation for cutting content. It requires three things:
a clear understanding of the desired performance outcome,
an honest assessment of the resources available,
and the professional confidence to tell stakeholders that not everything they want will make it into the programme.
Start with the problem, not the content. What does the person need to be able to do, or what does the organisation need to achieve, that they currently can’t? Gilbert’s Behavioural Engineering Model (1978) is useful here because it forces us to consider environmental factors, data, instruments, and incentives, before we assume the answer is training. If the performance gap is caused by unclear expectations or inadequate tools, no amount of carefully tiered training content will fix it. The triage lens applies to interventions broadly, not just to training design.
Once you’ve confirmed that there is a knowledge or capability gap worth addressing, map out everything that could plausibly be included in the response. Be comprehensive at this stage; the point is to get it all on the table before you start making decisions. Then apply the three questions to each element.
For knowledge: does someone need to know this to perform safely and effectively on day one? If yes, it’s Tier 1. Will knowing this improve their performance over time, even if they could manage without it initially? Tier 2. Is this interesting background that enriches understanding but won’t change what they do? Tier 3.
For skills: must someone be able to do this independently from the outset? Tier 1. Will this capability make them more efficient or more consistent once they’ve got the basics down? Tier 2. Is this an advanced or rare-use skill that most people will seldom need? Tier 3.
For problems: is this action essential to resolving the root cause? Tier 1. Will this address a contributing factor that amplifies the problem? Tier 2. Is this a peripheral improvement that might help but won’t make or break the outcome? Tier 3.
The temptation, and this is where professional courage becomes important, is to place almost everything into Tier 1. Stakeholders want comprehensive solutions, subject matter experts want their content included, and we want to feel we’ve been thorough. But if everything is Tier 1, nothing is, and you’re back to the original problem of spreading resources too thin.
Tier 3 is not the bin
One of the most important things to understand about this model is that Tier 3 content is not worthless. It’s content or activity that has some value, just not enough value to justify prioritising it over Tiers 1 and 2 when time and resources are constrained. In many cases, Tier 3 material can be made available as optional self-directed resources: further reading, reference guides, supplementary videos, communities of practice where people can explore topics at their own pace. The decision to deprioritise something is not the same as the decision to discard it.
One of the common objections to prioritisation is that it feels reductive, as if we’re deliberately limiting what people can learn or refusing to address parts of the problem. The opposite is closer to the truth. Prioritisation is what makes it possible for the critical elements to receive the attention, the design quality, and the facilitation time they deserve, rather than being squeezed into a programme alongside twenty other things that dilute the experience for everyone.
There’s a version of this that plays out with individual discretion too. Tier 3 items can often be left to the learner, the manager, or the team to pursue if and when it becomes relevant to them. This isn’t a failure of the programme; it’s a recognition that not everything needs to be centrally designed and delivered to have value, and that some things are better discovered through practice, conversation, and experience than through formal training.
The problem-focused lens
The three tiers become particularly powerful when applied not to individual roles but to organisational problems. When a team is trying to address, for example, a decline in customer satisfaction scores, the temptation is to launch a broad programme that covers communication skills, product knowledge, systems training, coaching for managers, and a revised incentive structure, all at once. The triage lens asks a harder question: of all these things, which ones must happen for the scores to improve?
Perhaps the root cause analysis reveals that the primary driver is a lack of product knowledge following a recent update. That’s Tier 1: people need to know the updated product specifications well enough to answer customer questions accurately. Tier 2 might be communication skills coaching that helps them handle difficult conversations more effectively, something that will improve the experience but isn’t the root cause. Tier 3 might be the revised incentive structure, which could have some long-term influence on engagement but is unlikely to shift customer satisfaction scores in the near term.
By sequencing the work this way, you ensure the most impactful intervention happens first, and you create a clear rationale for why some elements are being deferred rather than delivered simultaneously. It also gives you something to measure: if Tier 1 moves the needle, you’ve confirmed the root cause. If it doesn’t, you know to look elsewhere before investing further.
Note: It’s usually when exploring these kinds of examples that people say, “So how do I know what fits into tier one?” Well, being asked to figure that out should be the trigger point for you to be conducting analysis, having conversations, looking at data, and thinking deeply about what you understand about the linkage between activity and behaviour in the organisation and the KPIs they should be impacting.
Holding the line
The hardest part of this approach is maintaining it under pressure. Organisations love comprehensiveness; it feels safe, it looks impressive in a proposal document, and it distributes responsibility across enough activities that no single one can be blamed if the outcome doesn’t materialise. A focused intervention, by contrast, is exposed. If you say “this is the one thing that must happen,” and it doesn’t work, you’ve got nowhere to hide.
But that exposure is precisely what makes the approach more honest and more effective. It forces us to commit to a hypothesis about what will make the biggest difference, to test that hypothesis through implementation, and to learn from the result. This is how experiment-led practice works: you don’t try to change everything at once, you identify the highest-leverage intervention, you implement it, you measure its effect, and you iterate.
The triage lens is a tool for professional rigour in a context where rigour is often sacrificed in favour of perceived thoroughness. It doesn’t require a new framework, a new technology, or a new set of competencies. It requires the discipline to ask, for every element of every intervention, whether this is something that must happen, something that should happen, or something that would be nice, and to let the answer determine where our finite time and energy go.
We will never have enough resources to do everything. Accepting that, and choosing well, is the most important design decision we make.
References
Clegg, D. (1994). Case Method Fast-Track: A RAD Approach. Addison-Wesley. (PAID)
Gilbert, T. F. (1978). Human Competence: Engineering Worthy Performance. McGraw-Hill. (PAID)
Juran, J. M. (1951). Quality Control Handbook. McGraw-Hill.
Mager, R. F. and Pipe, P. (1997). Analyzing Performance Problems: Or, You Really Oughta Wanna. 3rd edn. Center for Effective Performance.
Pareto, V. (1896). Cours d’économie politique. F. Rouge. (This version is in the original French but can be downloaded and translated in an app of your choice. English translations can be found but are rarely free.)
Love the idea of an MVO....minimum viable outcome. Some clients, though, just won't get it.