KOL, HCP, and why the difference matters

Every key opinion leader is a healthcare professional. The reverse is not true. HCP is a factual description of somebody's profession; KOL is a judgement your organisation made about their influence, usually against criteria that were never written down anywhere.

The two terms get used interchangeably in conversation and they are not interchangeable at all. One is a fact that can be verified from a public register. The other is an assessment made internally, by people who mostly cannot say what it was based on.

That gap matters most at the moment a list gets made and somebody's name is on it or is not. It is also, quietly, one of the places where the measurement problem this sits inside becomes a governance problem instead.

What is an HCP?

A healthcare professional is anybody licensed to deliver care. Physicians, pharmacists, nurses, and depending on context dentists, physiotherapists and others. It is a factual category and registration is a matter of public record.

Inside Medical Affairs the abbreviation usually carries a narrower operational sense — a clinician the team may interact with — and that narrowing is where the first confusion enters. A pharmacist sitting on a formulary committee is a healthcare professional whether or not any MSL has ever met them, and whether or not they appear on anybody's list. The register does not care about your CRM.

What matters is what the term does not imply. Nothing about "HCP" says anything about seniority, influence, or interest in a particular product. It is a description of a profession, and its neutrality is the whole point of it. Both terms sit in the vocabulary Medical Affairs actually uses alongside the other words this field routinely overloads.

What is a KOL?

A key opinion leader is a healthcare professional whose views measurably shape the views of other healthcare professionals.

The word doing most of the work in that sentence is "measurably", and in practice it is rarely measured at all.

Part of the difficulty is that influence is not one thing. Three quite different properties get bundled under a single label.

Evidence influence. They publish, present, are cited, sit on guideline committees. This is the most visible kind and the only one verifiable entirely from public sources, which is why it dominates most KOL lists — not because it matters most, but because it is easiest to see.

Institutional influence. They chair a committee, run a department, control a protocol or shape a formulary decision. Entirely real, usually local, and often invisible from head office. A clinician with no publications at all can determine what an entire region prescribes.

Peer influence. Colleagues ask their opinion and act on it. The hardest of the three to observe and frequently the most consequential, because it operates in conversations nobody records.

Someone can be strong on one of these and absent on the other two. A single label collapses that distinction, and the collapse is where most of the trouble starts.

Why the criteria are rarely written down

Ask five people in a Medical Affairs team to define KOL and you will get five answers. That is not carelessness on their part. It is a governance gap, and there are reasons it persists.

The label is frequently inherited from commercial, where it existed for a different purpose and was maintained against different criteria. Nobody re-derived it when it crossed over; it simply came along.

Written criteria can be audited, and an auditable tier can be challenged — including, in principle, by the clinician it was applied to. Leaving the definition implicit avoids a conversation nobody wants to have.

And influence genuinely is hard to define. So the definition gets deferred, while the label goes on being applied. Deferral is a decision even when it does not feel like one.

Two things follow, and the second is the serious one.

Inconsistency. The same clinician is a top tier in one country and unlisted in another, and nobody can reconstruct why. Any cross-market view built on top of that inherits the incoherence.

Drift. An undefined label attracts meaning. "KOL" becomes shorthand for "person worth spending time on", which becomes "person likely to be supportive", and at that point a medical designation has turned into a commercial one without anybody deciding to make the change. It is the same failure as ranking people by how much they are worth, arriving through the side door.

That drift is the actual risk. Everything else is untidy; this one is a problem.

Why some medical teams have moved away from the term

Several reasons, and they are not variations of one another.

Compliance exposure. A system that ranks clinicians by likely commercial value is difficult to defend as non-promotional, no matter which function maintains it. The awkwardness is not that the ranking exists — it is that nobody can produce the criteria when asked.

The clinicians dislike it. Being labelled and ranked by a pharmaceutical company, generally without being told, is not an arrangement most clinicians would agree to if anybody asked them. That is worth sitting with rather than explaining away.

It does not do the job. A tier tells you somebody is important. It does not tell you what to discuss with them, what they already understand, what they asked last time, or what would actually be useful to them. It is a sorting label being asked to work as a plan, and it cannot. The same instinct produces the same error in measurement, which is why the measure is of the conversation, not the person.

Some teams have responded by renaming — "expert", "scientific leader", "external expert". Renaming on its own changes nothing. If the criteria remain unwritten, every problem above arrives intact under a new noun.

Which points at what the actual fix looks like, and it is not a better label.

What to use instead

Describe the influence rather than assigning the rank.

Four attributes, each recordable, each checkable, none requiring a judgement that nobody can reconstruct afterwards.

Evidence footprint. Publications, trial involvement, guideline participation, presentations. Public, verifiable, and it requires no opinion from anybody.

Institutional role. What they actually decide, and in which forum. Harder to gather than publication data and considerably more useful, because it is the thing that determines whether a conversation changes anything.

Scientific understanding. What they know about the topic and about the product — which changes over time and is the only attribute on this list that an interaction can actually move. It is also the one most worth tracking by product and by scientific topic rather than as a single figure.

Engagement history. What has been discussed, what was left unresolved, what they asked that never got answered. This is a knowledge position that survives a change of MSL, and it is the difference between a record and a list.

Four advantages follow from doing it this way, and they are worth stating explicitly because the tier system has inertia behind it.

Each attribute is a fact rather than a verdict, so disagreement is about evidence rather than about judgement. Each is auditable, which means it can be defended when somebody asks. Each is directly useful for planning the next conversation, which a tier label never is. And none of them can drift into a commercial ranking, because none of them is a ranking in the first place.

It is also a more honest account of what actually varies between clinicians. One can have a substantial evidence footprint and no institutional authority whatsoever. Another can have never published and control the treatment protocol for a region. A single scale puts those two people in an order, and the order is fiction.

If a tier is genuinely required for resource allocation — and sometimes it is — derive it from these attributes and write the derivation down. A number somebody can reconstruct is defensible. A number somebody assigned is not, and the difference will matter the first time anyone asks.

RocketMSL records what a clinician understands, what has been discussed and what is outstanding, per clinician and per topic — rather than a tier somebody assigned. See how engagement is recorded per clinician →