Biotech leaders ask better questions by switching from verification to discovery: fewer questions that confirm what they already believe, more that surface what the team has not said out loud. The change is small. The results are not.
Why this matters (and what most biotech executives get wrong)
Most biotech leaders were trained as scientists, and scientific questioning is precision questioning. You form a hypothesis, then you build a question designed to test it. That instinct is correct at the bench and wrong in a room full of people.
Precision questions close doors. At the bench, closing doors is progress. In a leadership meeting, a closed door means somebody's reservation never gets said out loud.
There is a second problem, and it has nothing to do with science. When the person who controls headcount asks a question, the room does not hear a question. It hears a preference. Ask "Are we still on track for the September readout?" and you have already told everyone which answer you want. Most of them will hand it to you.
For thirteen years my job was asking questions of people who did not want to answer them. I produced for ABC, NBC and Fox, which meant I asked strangers uncomfortable things on deadline, every day, and I had to come back with something usable or there was no segment. Here is what that taught me: the quality of an answer is set almost entirely by the question. And most of what people call questions are not questions at all. They are statements with a rising tone at the end.
The six-step process for asking better questions
1. Decide what the question is for before you open your mouth
Every question you ask does one of three jobs. Discovery: I do not know, tell me. Verification: I think I know, confirm it. Alignment: we need to land on the same page.
Nearly every failure I see is a leader using verification phrasing to do a discovery job. If the job is discovery, your hypothesis is not allowed inside the question. That is the whole rule. Name the job to yourself first, then build the sentence.
2. Open wide, then narrow
Producers call it the funnel. You start with the widest possible version of the question and narrow with each follow-up, instead of starting narrow and never finding out what you did not know to ask about.
Narrow: "Did we hit the enrollment target?" You get a yes or a no.
Wide: "Walk me through how enrollment came together this quarter." Now you get the two sites that stalled, the coordinator who left, and the reason the number is going to be a problem in March.
Same subject. One version has a floor and the other one does not.
3. Ask questions you genuinely do not know the answer to
Amy Edmondson's research on psychological safety puts this close to the center of what a leader can actually control: model curiosity, and ask genuine questions instead of leading ones. A genuine question creates a pull for people to share what they know. A leading question tells them what to say.
Here is the test. If you can predict the answer before you finish the sentence, it is not a question. It is a request for agreement.
While you are at it, kill the compound question. "Are we on track, and do we have CRO capacity, or should we push the timeline?" is three questions wearing one coat. People answer the last part and quietly drop the first two. I watch this happen in almost every leadership meeting I sit in on.
4. Follow up once more than feels comfortable
Alison Wood Brooks and Leslie John at Harvard Business School coded thousands of natural conversations and sorted questions into four types. Follow-up questions, the ones that simply ask for more, carried the most weight. They signal that you are listening and that you want to know more, and the people on the receiving end report feeling respected and heard (The Surprising Power of Questions, Harvard Business Review).
That matches what I see on camera. The first answer is the prepared answer. The second answer is the real one.
Four follow-ups worth memorizing:
- "What else?"
- "And then what happened?"
- "What makes you say that?"
- "What would have to be true for that to work?"
Ask "What else?" and then ask it again. The second one is where the useful material lives. It feels rude. It is not.
5. Let the silence do the work
Ask, then count to five before you say anything.
Most executives wait about a second and then rescue the room, either by answering their own question or by offering a multiple choice. Both are fatal. In an interview the pause is the single best tool you have, because people fill silence, and what they fill it with is almost always the thing they were deciding whether to say.
Five seconds feels like a minute. Do it anyway.
6. Ask the failure question
Cognitive psychologist Gary Klein's premortem flips the framing. Instead of asking what could go wrong, you tell the room the patient is already dead. "It is eighteen months from now. This program failed. Write down why." Klein's work found the exercise improves people's ability to identify reasons for future failure by roughly 30 percent, because it gives them permission to say things that would sound disloyal during ordinary planning (Performing a Project Premortem, Harvard Business Review).
Two rules make it work in a biotech setting. Everyone writes silently and individually before anybody talks, so the loudest person in the room does not set the frame. And you go around the table, so nobody gets to pass.
The mistake most biotech executives make
They ask questions to demonstrate rigor instead of to gather information.
In a room full of scientists, a hard question is also a status move, and everybody knows it. Your team can tell the difference between a leader who wants to understand something and a leader who wants credit for catching something. When your questions are performing rigor, people stop volunteering the exact information you need, because volunteering it hands you ammunition.
The tell is the closing question. "Any concerns?" is not a question. It is a request for silence, and it works every single time. It asks people to raise a hand, in front of their peers, and be the one slowing everything down.
Try "What is the weakest part of this?" instead. It presumes there is one, so naming it costs nothing.
Take the question you normally use to close a meeting and replace it with "What is the weakest part of this?" Then count to five before you speak. That is the entire assignment, and it will tell you more about your team's information flow than a survey will.
Case study: three versions of the same sentence
I spent two days in Chicago with four senior physicians getting ready for a conference where they were the headline speakers and journalists would be in the audience. They booked me for media training. By the end of the first exercise they had added a second day for presentation training.
The problem showed up in round one. They were deep in the weeds. The language was accurate and it would have lost a general audience inside of thirty seconds. The fix was never about dumbing anything down. It was about building stories and analogies that made their life's work land with anyone in the room.
But the part that belongs in this article happened before any of that.
The most senior physician asked the others whether the core material was clear. Everyone said yes. That question got a nod, and the nod meant nothing.
So I asked a different one. I asked each of them, one at a time, to say the same content back to me in a single sentence a smart sixteen-year-old would repeat at dinner. Three noticeably different versions came back. Not shades of one idea. Different claims.
That gap had been sitting in the room the whole time. One question buried it and the other found it in about four minutes. This is what I mean when I say the question sets the answer.
The question that gets a nod vs the question that gets information
| The question that gets a nod | The question that gets information | Why the second one works |
|---|---|---|
| Any concerns? | What is the weakest part of this plan? | Presumes a weak part exists, so naming it is not disloyalty |
| Are we on track? | Walk me through what has changed since the last update. | No hypothesis inside the question, so there is nothing to agree with |
| Does everyone understand? | Say it back to me in one sentence. | Tests comprehension instead of asking people to self-report it |
| Do you see this as a risk? | It is a year from now and this failed. What happened? | Removes the social cost of being the pessimist in the room |
| Are you confident in the data? | What would change your mind about this result? | Surfaces the boundary of the finding, which is the useful part |
What to do next
Pick one meeting this week and change one question. Nothing else. Leaders who try to overhaul how they run a room end up back where they started by Thursday.
If your leadership team is heading into a data readout, an investor meeting, or a regulatory conversation, the asking and the answering are the same skill pointed in opposite directions. Our executive communication coaching covers both sides. For teams getting ready for a financing conversation, investor pitch training is the closer fit.
Related reading: Is It Safe Is Not a Yes or No Question on the answering side of the same problem, and When the Org Chart Made Scientists the Spokespeople on what happens when technical experts inherit the questions.
Want a session built around your team's actual meetings? Get a quick quote.