Healthcare
Five behavioural biases quietly shaping pharma brand decisions
Pharma brand teams like to think of themselves as analytical. Sometimes too analytical. But every brand plan, launch decision, and investment trade-off is being made by humans — under pressure, with partial information, in rooms with power dynamics.
Which means the decisions are subject to exactly the same cognitive biases as anyone else's. Here are five we see most often.
1. Availability bias
The last anecdote in the room wins. A story from one customer advisory board will shape a global strategy more than a tracker with a thousand respondents. Not because the anecdote is better data, but because it's more vivid.
What to do about it: ask, out loud, what story is currently doing the heavy lifting in the decision. Then ask what you'd need to believe about the wider population for that story to be representative.
2. Sunk-cost bias on agency relationships
Teams stay with an agency long past the point of diminishing returns because the thought of re-briefing is exhausting. The cost of the switch gets priced in; the cost of the status quo doesn't.
What to do about it: every two years, write down what you'd hire the agency to do today if you were starting fresh. If the answer isn't what they're actually doing, you have a decision to make.
3. Loss aversion in message testing
The strongest message usually tests worst because it polarises. Teams chase the one that everyone agrees is "fine" and end up with something that doesn't offend but doesn't shift anything.
What to do about it: track both top-box appeal and bottom-box rejection. A message that 40% love and 20% reject is almost always more commercially useful than one that 65% find pleasant.
4. Planning fallacy on launch timelines
Every launch plan assumes smooth execution. No launch has ever had smooth execution. The gap between the plan and reality eats most of the contingency before week six.
What to do about it: build the plan once. Then ask the brand team what could realistically slip, and by how much. Add it to the plan. Then ask again. You will still be short.
5. Authority bias in KOL consensus
When two KOLs out of eight hold a strong view, that view often becomes "what KOLs think". Especially if the two are more senior, more published, or more charismatic. The other six quietly disagree without disagreeing.
What to do about it: ask for the view anonymously, or at least ask each KOL individually before the group call. You'll get a different picture.
The meta-bias
The most dangerous bias in pharma brand teams is the one that says "we're a data-driven organisation, so we're probably fine." Being data-driven doesn't protect you from this. It just gives you more impressive-looking ways to justify the decision you were already going to make.
The work is to notice when the reasoning is running backwards from a conclusion, and create small structural checks that make it harder to do that.