Ten Questions Your Audience Will Ask After Your Presentation

You can prepare for a presentation perfectly — and still get taken down by what comes after: the questions. You have your slides under control, not the questions, and it's exactly the answer "I'll have to look into that" at the crucial moment that decides the impression left by the whole half hour. The trick is simple: questions are predictable. Audiences ask surprisingly predictable things — about money, about risk, about "why not do it differently" — and AI can generate from your materials the very list you'd otherwise hear for the first time live.

How to do it

1. Paste your outline, slides, or presentation script into the chat and describe the audience.
2. Use this prompt:

Here's my presentation. Audience: [e.g. company leadership,
15 minutes, deciding on next year's budget].

Generate the 10 questions most likely to come up afterward:
- 3 curious ones (want an extra detail),
- 3 skeptical ones (don't believe a number or a conclusion),
- 2 about money and return on investment,
- 2 uncomfortable personal ones ("why didn't you finish
  this already this year?").

For each question, note whether the materials already
answer it (and where), or whether it's MISSING. Don't
answer any of them for me.

3. Questions marked MISSING are your priority: prepare an answer, add it to a backup slide, or line up an honest "we'll have that number by Friday."
4. The day before, do a live dress rehearsal: "Ask me the questions one at a time, I'll answer, then tell me what my answer was missing."

A typical scenario

Project manager Klára is pitching leadership on extending a project by a quarter. Her presentation is solid, but she has the questions generated anyway — and two on the list make her freeze: "What did you do with the buffer you got in March?" and "How much will a month's delay cost the customers?" She has an answer to the first in her head, but not in numbers; to the second, nothing at all.

The evening goes to those two questions, not to polishing slides. In the meeting, the first question really does come up, almost word for word — and Klára opens the backup slide with the buffer breakdown. The extension gets approved. Not because she had a prettier deck, but because it wasn't the first time she was confronted with it live.

What you get out of it

Questions stop being a lottery. The generated list will never hit everything, but it hits the center of mass — money, risk, disputed numbers — and, above all, shows you the holes in your materials while they're still fixable. That "don't answer any of them for me" in the prompt is deliberate: the answers have to be yours, or the first follow-up question sinks you. If you want to go further and have the content itself stress-tested, there's the tip let AI look for holes in your proposal.

Pro tip

After the presentation, note down the questions that actually came up, and next time paste them into the prompt as examples: "this is how my audience asks questions." After three presentations, the model generates questions strikingly close to the ones that actually get asked — every audience has its own signature, and AI learns it from examples.