Key takeaways
Knowledge and skill are separate dimensions, so passing a demanding knowledge assessment does not necessarily demonstrate job readiness.
- Bloom’s taxonomy measures increasingly sophisticated engagement with knowledge, not whether candidates can perform a task.
- Skill becomes visible when candidates must act, adapt to feedback, and manage consequences.
- Static simulations, adaptive simulations, supervised practice, and job performance offer increasing levels of fidelity.
By Whitney Coggeshall
When asked whether they're assessing skills, many professional education programs point to Bloom's taxonomy. Specifically, they point to the levels above simple recall, including application, analysis, evaluation, and creation, and argue that assessing at these levels means assessing skills. It's a reasonable-sounding claim, and the language makes it easy to believe. "Application" in particular implies doing something, which is presumably what skills are about.
But when a candidate demonstrates "application" on an assessment, what are they actually doing? However sophisticated the reasoning required, the candidate is still working with words on a page. They're reasoning about action, not performing it.
That distinction turns out to matter more than we tend to acknowledge, and part of the reason it gets overlooked is a deeper assumption most of us share that knowledge and skill are just different points on the same scale. Study more, go deeper, and understanding eventually becomes competence. It's an intuitive model, but it's also wrong.
Knowledge and skill are not on the same continuum
They're different dimensions, and like any two dimensions they can be related. Knowledge often supports skill, and with enough real practice it can even turn into it. But the relationship isn't automatic, and that's exactly where the trouble starts. You can have deep knowledge that never became skill, simply because it was never practiced under conditions that demanded performance. The clearest way to see the gap is to look at what happens when the two come apart.
Consider four people.
The Communication Scholar has spent years studying patient communication. She can describe the principles of empathetic listening, explain frameworks for delivering difficult diagnoses, and write thoughtfully about what good bedside manner looks like. But put her in a room with a frightened patient and she freezes, over-explains, or reaches for clinical language when plain words are what's needed. High knowledge, low skill.
The Intuitive Designer has spent years designing training programs. She knows intuitively what works, including how to sequence content, when to use practice, and how to keep learners engaged, and her programs consistently produce results. Ask her to name the learning science principles behind her design decisions and she'll struggle, because her knowledge is tacit, built from experience rather than study. High skill, lower formal knowledge.
The third is a Novice, low on both.
The Expert has spent years both doing the work and studying it seriously. She has the conceptual foundation to understand what she's doing and the practiced judgment that only comes from having done it many times, under real conditions, with real stakes. High on both. And notice that it wasn't deeper knowledge alone that got her there. It was deliberate practice.
These aren't edge cases. They're familiar, because knowledge and skill are genuinely separate things.
Where Bloom's actually lives
Bloom's taxonomy is a well-designed and useful framework. It helps us write better learning objectives, design more demanding assessments, and move learners beyond simple recall, and none of that is in question here.
But Bloom's sits entirely within the cognitive domain. Every level, from Remember to Create, describes a more sophisticated way of engaging with knowledge declaratively, and all of it happens in language, on paper, one step removed from action.
It's worth heading off an objection here. The revised taxonomy includes a category called procedural knowledge, which sounds like it finally covers skill. But procedural knowledge, as the framework defines it, is knowledge of how to do something, the steps, the methods, the sense of when to use them, and knowing all of that is still different from being able to carry it out when it counts. You can write an assessment at the very top of Bloom's that asks a candidate to work through a procedure on paper and never once requires them to perform. The level tells you how demanding the thinking is. It doesn't tell you whether anyone did the thing.
You can reach the top of Bloom's and still be in the bottom half of the skill axis, which isn't a criticism of the framework so much as an honest account of what it was designed to do.
The threshold between knowing and doing
The line between knowledge and skill isn't about cognitive demand. It's about whether someone is describing an action or performing one.
Take prompting a generative AI model. You can write a thoughtful essay on prompting strategy, the value of specificity, chain-of-thought reasoning, what makes a prompt fail, and it can reflect real insight. But actual prompting skill is iterative. It lives in reading the output, seeing what went wrong, adjusting, and deciding in the moment what to try next, and no essay captures that.
Or take building a financial model. A candidate can answer every conceptual question correctly, when to use which valuation method, how to treat different line items, what drives the output, and still sit down with a messy dataset and a deadline and not know where to start. The knowing and the doing are different, and the difference shows up the moment you have to do it.
This is what crossing the threshold means. The task pushes back, and the environment responds to what you actually do rather than to what you say you would do.
Once you've crossed the threshold
Acknowledging the threshold doesn't mean all skills assessment looks the same. Once you're in the procedural domain, there's a genuine continuum from contrived to real, running roughly from knowledge assessments and case study essays through static simulations, adaptive simulations, and supervised practice, all the way to actual job performance.
A static simulation presents a realistic scenario but follows a fixed path regardless of what the candidate does. It asks candidates to make decisions rather than describe them, which is a meaningful step forward.
An adaptive simulation goes further. When well designed, it responds to what the candidate actually does, forces iteration, and exposes poor decisions through consequences rather than evaluator feedback, so a candidate who makes a bad call has to deal with what happens next. That's closer to genuine skill demonstration, but even a well-designed simulation is still contrived. The stakes aren't real, the context is controlled, and the candidate knows they're being assessed, so it sits somewhere in the middle of the fidelity continuum rather than at the end.
The more pressing issue is that most professional credentialing programs aren't yet on this continuum at all, not because their assessments are insufficiently rigorous, but because they haven't crossed the threshold from declarative to procedural. The candidate is always describing, never doing.
What this means for how we credential
When a professional assessment asks candidates to analyze a scenario, evaluate competing recommendations, or construct a reasoned argument, they're doing something genuinely difficult, and that difficulty shouldn't be dismissed. The ability to reason carefully at high levels of cognitive demand is valuable and worth measuring.
But the candidate is still doing one thing: writing about/selecting what they would do. The action remains one step removed, and describing the right action isn't the same as being able to reliably perform it under real conditions when it counts.
That gap between knowing and doing is exactly where job readiness lives, and most of our credentials say very little about which side of it a candidate is on.
Three questions worth asking
For anyone designing or evaluating a professional education program, three questions are worth sitting with.
- Could a candidate answer this correctly without ever having done the thing? If yes, we're assessing knowledge, not skill.
- Does the assessment require the candidate to actually perform, and does the task push back when they get it wrong? If a candidate can succeed by describing the right answer rather than producing one, we haven't crossed the threshold.
- If this candidate passed, would we trust them to perform on day one? If the honest answer is "not really," the credential tells us something valuable about what they know, but it doesn't tell us what they can do.
Knowledge and skill are both worth building and both worth assessing, but they're not the same thing, and the field won't close the job-readiness gap until we treat them accordingly.
You may also be interested in
Explore our programs and certificates
CFA Institute offers a diverse range of programs and certificates designed to meet the needs of finance professionals across various career stages and specializations.