Exercise more. Check email less. Finally unsubscribe from that streaming service we haven’t watched in 6 months.
That gap between what we intend to do and what we actually do does not disappear in a clinical setting. What HCPs say they intend to do and what happens in practice do not always line up.
Because the most important demand question isn’t simply “Will they?”
It’s “What will make them?”
Here are some of our best practices for delivering a more accurate and actionable demand assessment:
1. There is no single “right” demand methodology
One of the easiest traps in demand research is starting with the technique: For instance, should we run a conjoint?
The better starting point is: What uncertainty are we trying to resolve? Demand questions change across an asset’s lifecycle, and the methodology should change with them.
Are we estimating uptake for a largely defined target product profile (TPP)? A TPP allocation can provide a direct read on potential uptake.
Are we deciding which attributes of an uncertain future profile matter most, and which trade-offs HCPs will accept? Conjoint can help explore those choices.
Or are we asking which patients are most likely to be considered for treatment? Patient chart audits can help show how potential demand varies across different patient profiles.
2. It helps to move from stated preference to realistic choice
Choosing the right methodology is only the beginning. The next challenge is recreating the context in which decisions actually happen.
HCPs do not make prescribing decisions in a vacuum. Existing treatment experience matters. So do patient characteristics, efficacy and safety trade-offs, dosing, familiarity and the alternatives available at the moment a decision is made.
Well-designed demand research should reflect those trade-offs.
Conjoint, for example, can create a decision environment in which HCPs respond to different treatment configurations and make choices among available options. A fixed TPP approach asks a different question, allowing HCPs to evaluate a complete product profile and estimate which patients could initiate or switch.
Neither approach is fundamentally superior. They solve different problems.
What matters is getting respondents close enough to the real choice that the resulting estimate becomes commercially meaningful.
3. The number is only half the answer
A demand estimate of 18%, 27%, or 35% can look reassuringly precise.
But the more valuable question is usually why?
Which product characteristics are driving demand? Which patient types offer the greatest headroom? What existing behaviors protect an incumbent? What could a competitor do to change the equation? Which elements of the future profile create upside or risk?
Commercial potential should not stop at identifying an opportunity. We need to understand its size and accessibility, what drives adoption, and what would be required to capture it.
A demand number tells you what could happen. A strong demand model helps you understand what would need to happen to make it real.
4. Calibrating the say-do gap improves accuracy
Even the best-designed choice exercise cannot fully recreate real-world behavior. That’s where behavioral science can strengthen the demand estimate.
We look beyond stated uptake to understand readiness to change: the difference between finding a new treatment appealing and being motivated to change established prescribing behavior.
An HCP may respond positively to a new treatment but have little motivation to change current prescribing habits. Another may see significant unmet need with existing options and be much more receptive to something new.
The same stated demand can therefore represent very different adoption potential.
Behavioral intent can provide an important calibration input. Alongside stated uptake, we examine intent to change prescribing behavior, satisfaction with current options and perceived unmet need. Taken together, these help distinguish active readiness to change from more passive interest and identify where stated demand needs more cautious interpretation. They inform the estimate; they do not prove that adoption will follow.
For example: two HCPs might state the same expected uptake for a new treatment. One sees substantial unmet need and is motivated to change; the other is satisfied with current options and sees little reason to switch. The demand estimate looks the same, but the research points to different barriers – and different conditions needed to convert that interest into adoption.
Being interested is not the same as being willing to change. And being willing to change is not the same as actually changing.
A new treatment option must feel relevant enough to warrant consideration, deliver enough clinical value to justify the trade-offs, and overcome the habits, barriers and friction standing between consideration and behavior change.
That is the gap demand assessment needs to navigate.
Using demand assessments to make commercial decisions
Ultimately, demand assessment insights should not end with a percentage on a slide.
For teams making development, forecasting, positioning or launch decisions, knowing the size of potential demand is useful. Understanding where it comes from, how robust it is and what could change it is what makes the research actionable.
The real value is in the questions the demand estimate helps us answer:
Where is demand strongest? What is driving it? What could unlock more of it? What could put it at risk? And what would need to change for stated interest to become real-world adoption?
That moves the conversation beyond “How much share could we get?” Instead, we can ask: “Where will growth come from, what will drive it, and what do we need to do to capture it?”
At Branding Science, that is the role we see for demand assessment: connecting potential demand with the decisions and conditions that could turn it into adoption. Because the goal should not simply be to predict what people say they will do.
By understanding the say-do gap, we can move beyond stated intent toward more accurate, actionable demand assessments.