How MedComms agencies prove ROI to pharma clients in 2026: a practical guide

Posted in Uncategorized

Pharma now wants behaviour-change evidence, not impressions, not attendance, not completion. That means measuring at Moore’s Level 4 (competence) or Level 5 (performance) as a minimum, with the proof captured inside every educational asset rather than bolted on with a post-event survey. Merck’s published medical-education grant criteria already require it. So do Insmed’s. The agencies still reporting reach and frequency are getting cut. The agencies showing distance travelled are winning the next brief.

---

So the question every MedComms head is being asked right now is some version of this one: “Did it work?” Procurement asks it. Medical affairs asks it. The brand director asks it. And most agency reports answer a different question: they show what was sent, what was opened, who turned up. That’s not the same thing. That’s the gap that’s costing renewals.

I’ve been on the board of publishing groups, run platforms for medical education, and watched the same thing happen in different rooms, beautiful content rotting away on bookshelves while the people who paid for it can’t tell whether anyone’s behaviour shifted. Pharma noticed. They’re rewriting the briefs.

This guide is what we’ve learned from inside that shift: what proof looks like in 2026, where the bar has moved to, and what MedComms agencies need to put in their next response document.

What pharma’s actually asking for now

The shift is already in writing. Merck Global Medical Unit’s published Call for Grant Applications (CFG-23-001) states it plainly:

“It is a requirement that medical education providers evaluate the educational outcomes with a learning effectiveness measurement system… A minimum of Level 4, preferably Level 5, on the Moore’s scale is expected.”

That isn’t aspirational language. It’s the gate. Insmed’s Call for Grants uses the same minimum. Bristol Myers Squibb’s guidance for IME projects sits in the same place.

So the brief from procurement isn’t “give us a great programme.” It’s “give us a programme that produces Level 4 or Level 5 outcomes data, and tell us in your proposal how you’re going to measure it.” Agencies that can’t structure an answer to that question are ruled out before the creative is even read. That’s a quiet, unsexy procurement filter, but it’s where margin gets won and lost in 2026.

Why most current MedComms reports miss

The honest answer: most reports were never designed to measure performance. They were designed to measure delivery. So you get traditional reach and frequency metrics: page impressions, attendance, completion rates, sentiment scores, and a closing slide that says “high engagement.” Pharma reads that slide and asks the only question that matters: did anyone change what they prescribe, recommend, or do?

Tiffany Pela, US Gastroenterology Medical Director at Sanofi, put it like this: leadership expects KPIs that link activity to impact, but reach-and-frequency thinking is “inappropriate for Medical Affairs” and compromises scientific credibility. ZS’s 2026 Medical Affairs trends report finds 69% of KOLs now expect data on real-world evidence and patient-reported outcomes from MSLs, the same pressure is now landing on the educational side.

That’s the heart of it. The buyers want behaviour change. The current toolset captures attention. Two highly polished stones glancing past each other.

Moore’s Levels, the gold standard, in one minute

The framework everyone’s pointing at comes from Donald E. Moore Jr at Vanderbilt, originally published in the Journal of Continuing Education in the Health Professions in 2009 (DOI: 10.1002/chp.20001). It maps medical education outcomes across seven levels:

• Level 1: Participation
• Level 2: Satisfaction
• Level 3a: Declarative knowledge (“knows what”)
• Level 3b: Procedural knowledge (“knows how”)
• Level 4: Competence (“shows how”)
• Level 5: Performance (“does”)
• Level 6: Patient health
• Level 7: Community health

Most MedComms reports stop at Level 1 or 2. Pharma is now writing briefs at Level 4 or 5. The ACEhp Outcomes Pyramid sits in the same place. So does Alcon’s public CME outcomes guidance.

That’s the bar. Five levels above where most reporting currently lands.

The distance-travelled approach: proof, built in not bolted on

The shift that actually works isn’t a better post-event survey. It’s measurement built into the asset itself. So before any HCP touches the content, you capture a baseline: confidence on a topic, what they currently recommend, how they’d act in a clinical scenario. The same prompts re-run after the engagement. The delta is the distance travelled. Every interaction. Every HCP. Every cell.

That’s how we built Yoke. The numbers don’t come from a separate evaluation tool, they come from the same content the HCP was using to learn, and they’re computed per programme from that cohort’s own baseline-to-endpoint data. Built in, not bolted on.

The point isn’t the platform. The point is the principle. Pharmaceutical Executive recently described it as connecting “the dots between programme exposure, engagement, and behaviour at the physician level, with confidence.” Improvado’s pharma analytics review describes the same shift toward outcomes-linked measurement. PharmaNow calls it value-based communication. Different vocabulary, same direction of travel.

A practical four-step plan for the next pitch

If you’re sitting in a MedComms agency and the next RFP lands tomorrow, here’s the structure that turns a fee-compressed pitch into a renewal-locking one:

1. Read the funder’s outcomes language first, not the creative brief. Open the procurement document. Search for “Moore.” Search for “Level 4.” If those words are there, build the pitch backward from the measurement plan. The creative gets shorter. The proof section gets longer.
2. Pre-register your KPIs. State which Moore’s level you’re aiming at, what the baseline measure is, and how the post-measure will be captured. If you can’t answer that in one paragraph, you don’t have a measurement plan, you have a hope.
3. Build the proof into the asset. Pre/post confidence prompts. Scenario decisions. Behaviour intent. Inside the cell, not in a SurveyMonkey afterwards. Pharma can read engagement-as-measurement on the same page they read the content.
4. Report distance travelled, not delivery. A one-page summary that opens with “+X percentage points across [N] HCPs on [topic], measured against baseline.” Then the rest. Procurement reads the first line. Make it the answer to “did it work?”

That’s the structure. Nothing fancy. Just the discipline of measuring in the right place, at the right moment, against the right scale.

The bit nobody likes saying out loud

Pharma budgets are tight. Procurement is hunting for marginal cost-out everywhere. The agencies that will be cut first are the ones answering the wrong question, beautifully, expensively, and on time. The agencies that survive the next twelve months are the ones who can hold a line in a pitch and say: “Here’s what we measured, here’s what changed, here’s the proof.” Bare. Specific. With a number.

That’s not a content question anymore. It’s a measurement question. The agencies that get that earliest get the next brief.

FAQ: what MedComms buyers actually ask

What does pharma mean by “outcomes” in 2026?

Behaviour change at Moore’s Level 4 (competence) or Level 5 (performance): what the HCP can do or actually does after the activity, not what they remember or rated highly. Merck and Insmed both write Level 4 minimum into their grant criteria.

Why aren’t completion rates and attendance enough anymore?

Because they answer the wrong question. They prove delivery, not impact. Sanofi’s gastroenterology medical lead calls reach-and-frequency thinking “inappropriate for Medical Affairs”, pharma boards now ask whether prescribing or recommendation behaviour shifted.

What’s Moore’s Level 5 in practice?

Performance: what the HCP actually does in clinical practice after the activity. Captured through behaviour-intent scenarios, prescription pattern data, or follow-up performance prompts at 4 to 6 weeks. Moore’s original framework is the canonical reference.

Can our agency hit Level 5 without a measurement platform?

You can hit Level 4 with well-designed pre/post questioning. Level 5 typically needs a measurement layer that follows the HCP past the event: a 4 to 6 week behaviour follow-up loop captured against baseline. That’s harder to do in a one-off PDF or live event without infrastructure.

How do we explain “distance travelled” to a brand team that thinks in reach?

One sentence: “We’re moving from how many people we reached to how far each person moved.” Pair it with a single metric: confidence delta, recommendation delta, scenario-decision delta. Make the proof unit a number, not a slide.

Will procurement actually pay more for outcomes evidence?

Not always more, but they’ll pay you over a competitor who can’t answer the outcomes question. Pharmaceutical Executive’s analysis is consistent with the procurement reality: outcomes data turns a tactical brief into a measurement contract. That’s where retention sits.

What’s the fastest way to retrofit measurement onto an existing asset?

Add three things: a baseline confidence/behaviour prompt at the start, the same prompts at the end, and one scenario-based decision question. Reported as a delta against baseline, that single change can take you from Level 2 to Level 4 inside a week.

How do we report it back to pharma?

A one-page proof artefact, opening line first: “+X percentage points distance travelled across [N] HCPs on [topic].” Then the breakdown by sub-cohort, the scenario-decision deltas, and the methodology footnote. The same artefact every quarter, same structure, so trend lines build over the contract.

---

Matt Davies is the founder of Yoke Health, a medical-education platform for pharma, biotech, and MedComms agencies. Yoke captures behaviour-change evidence inside every educational interaction, so the proof arrives with the content, not after it.

Want to see the proof artefact pharma actually wants? Get in touch.

 

---