Healthcare has always been deeply human. The decisions shaping it deserve to be too.

Here is how most healthcare teams test a message.

They write it. They pull eight people into a focus group. They ask what they think. Then they launch.

No. That isn’t research. That is a polite stranger telling you what they think you want to hear. And then you bet a clinical trial, a drug launch, or a $40M member engagement program on it.

This is the part nobody talks about.

The 7pm problem

Your team writes a refill reminder at 2pm on a Tuesday.

The patient reads it at 7pm after a 10-hour shift, while cooking dinner, while their kid is asking about homework.

Those are two different messages.

The second one is the one that actually counts. And your focus group will never show it to you. Because the person in the focus group knows they are in a focus group.

So they perform. They give you the rational answer. The one they would give if they were the person their doctor thinks they are.

They are not that person at 7pm.

What we actually do

We build synthetic populations.

Thousands of AI personas, each with real health literacy, real trust patterns, real daily-life constraints, and real psychological responses to being treated like a patient.

A 71-year-old retired teacher who reads at a 9th-grade level, manages three chronic conditions, and doesn’t trust pharma.

A 34-year-old caregiver running her mom’s meds and her kid’s pickup in the same hour.

A cardiologist with 14 minutes between patients and a strong opinion about the last rep who wasted his time.

Then we run your message against them. Before you ship. Before you spend. Before the first real patient ever sees it.

Here is where this changes outcomes.

1. Clinical trial recruitment messages

80% of trials miss enrollment timelines. Every day of delay costs $600K to $8M.

The default response is to spend more on recruitment. The better response is to check whether the recruitment line even reaches the people the trial was designed for.

We keep seeing the same pattern.

Help advance science converts highly educated patients who already volunteer for research. Great for the easy tier.

That same line collapses with the underrepresented communities most trials urgently need.

For those patients, get access to treatment you cannot get anywhere else converts 3x better.

Same trial. Same protocol. Different frame.

If the framing doesn’t match the segment, no recruitment budget will rescue it.

2. Adherence nudges

The people designing adherence programs have usually never lived with a chronic condition.

So the program they ship is well-intentioned and quietly infantilizing.

Push notifications work for newly diagnosed patients. They want the reminder.

Those same notifications disengage patients who have managed the condition for ten years. They read as nagging. As the 400th reminder that they are sick.

That second group doesn’t respond to apps. They respond to a pharmacist texting them by name.

Not because the content is better. Because a human relationship is carried by the channel.

The message is the channel. You cannot test one without the other, and most programs never test both.

3. MSL and HCP messaging

A Medical Science Liaison gets three minutes of a physician’s attention.

One bad opening line burns the whole visit.

We split synthetic HCP populations by how they actually prescribe. Evidence-conservative. Early-adopter. Patient-preference-driven. Guideline-adherent. Then we run your talking points against each.

What we find most often: community physicians don’t care about pivotal trial data in the first 30 seconds.

They care about which of their patients will actually benefit, and which will show up back in their office next week with adverse events they have no support staff to manage.

Academic physicians want the data. Community physicians want the selection criteria.

If 60% of your target list is community, and your sales deck leads with the data, you are wasting the first 30 seconds of every visit.

4. Wellness and preventive care content

Sending the same wellness content to every member isn’t personalization. It is noise.

Clinical content (your A1C means…) works for one cohort.

Lifestyle content (5 meals that help manage blood sugar) works for another.

Community content (people like you found…) works for a third.

Send the wrong one to the wrong person and you don’t just lose engagement. You actively alienate your highest-risk members.

A cost-conscious member reads catch it early as scare-tactic marketing.

An anxious member reads $0 under your plan as dismissive.

Both unsubscribe. Neither books the screening.

5. Pricing and access messaging

A pricing decision is never just a pricing decision.

It is a message to patients. A message to prescribers. A message to payers. A message to advocacy groups. A message to journalists.

Each one reads the signal differently.

You can negotiate a lower price and win on affordability. And simultaneously tell early-adopter physicians that the drug is older generation, which quietly reduces new starts among the exact prescribers you were counting on.

Economic models don’t see this. Behavioral simulations do.

We simulate pricing decisions across every stakeholder at once, with real incentives, real trust levels, real political pressure.

The output isn’t a revenue curve. It is a behavioral cascade.

Who abandons therapy in Month 4 when they hit the coverage gap. Which physicians shift prescribing on signal, not cost. Which advocacy groups amplify the story, and which bury it.

6. Omnichannel sequencing

Most healthcare orgs treat channels as independent.

Email owns email. SMS owns SMS. Field force owns field force. Each team optimizes their own number.

The audience experiences the combination. And the combination is usually redundant, out of order, or contradictory.

Evidence-conservative physicians need three data touchpoints before an MSL visit is productive.

Early-adopter physicians are ready to prescribe after one peer conversation. A second data push reads as condescending.

A one-size-fits-all call plan wastes roughly 40% of your field capacity on mis-sequenced engagement.

Sequencing is a message too. Getting it wrong says more than the content.

Why this matters

Traditional research answers the wrong question. Too slowly. At too small a scale. With observer bias baked in.

We have watched teams spend $2M on a congress presence to learn what a simulation would have told them in two weeks.

We have watched trials miss enrollment by six months because the recruitment line was tuned for the wrong demographic.

In healthcare, a misaligned message isn’t wasted spend.

It is a patient dropping out of a trial that would have helped them.

It is a member unsubscribing from the exact program designed to save their life.

It is a physician forming a first impression no follow-up will ever reverse.

Test the message against the audience you actually have. Not the one you assume.

One call

If you have a trial about to open, a drug about to launch, a digital program about to roll out, or a pricing decision about to go public, we would welcome a 30-minute call.

No pitch deck. One real decision. One focused simulation. Delivered in weeks, not quarters.

Start a conversation →