Shows up to meetings with nothing sharp to ask. Ten minutes turns a skimmed pre-read into two questions that make you look prepared instead of passive.
| Prep Time | About 10 minutes using the pre read and prompt method described in the article |
|---|---|
| Weekly Meeting Waste | US employees lose roughly 5 hours a week to unproductive meetings, per Flowtrace’s 2026 meeting statistics roundup |
| AI Adoption At Work | 41% of US workers reported using generative AI for work tasks as of November 2025, per Federal Reserve FEDS Notes |
| Questions And Likability | Asking thoughtful follow up questions increases likability and perceived competence, per Harvard Business Review’s 2018 research summary |
1. Grab the pre-read the moment it lands
As soon as someone sends an agenda, deck, or proposal, copy the text instead of letting it sit unread until the meeting starts. Do this right away, not five minutes before the call when you’re rushed. If it’s a PDF or slide deck, copy the text or paste a screenshot into your AI assistant.
2. Use one repeatable prompt
Paste the document in and ask: ‘Give me 3 clarifying questions, 2 questions that show I’ve thought ahead, and 1 question that surfaces a risk or gap in this document. Also list anything vague, missing a number, or missing a named owner.’ Save this exact wording somewhere so you can reuse it before every meeting.
3. Pick two, not ten
AI will hand you six or seven questions. Choose the two sharpest ones, usually one clarifying question and the risk question, and leave the rest as backup. Two strong questions beat a list you’ll never get through.
4. Open with the risk question
If the meeting allows it, ask the risk or gap question early instead of at the end. It signals you read the material closely and sets a more serious tone for the rest of the conversation.
5. Keep the gap list as your follow-up
Hold onto whatever the AI flagged as missing, an unnamed owner, an unclear number, a vague date. If the meeting doesn’t address it, you already have your follow-up email half written.
A worked example
A vendor sends a renewal proposal an hour before tomorrow’s call.
Before
Annual license renewal, 12% increase, includes priority support, implementation timeline TBD.
After
Clarifying: What’s driving the 12% increase, inflation or added features? Is it a flat increase or compounded on current list price? What does priority support include that last year’s plan didn’t? Sharp: If we commit to a two-year term now, does the increase lock in or apply again next year? What happens to pricing if our usage drops below the current tier? Risk: There’s no implementation timeline for the new support tier. Who owns getting that on paper before we sign? Gaps flagged: timeline TBD, no cancellation terms mentioned, no named account owner.
Three ways to bend it
Before a job interview
Paste the job description and the company’s about page, then ask for questions to ask the interviewer that show you understand the role, not generic ones like ‘what’s the culture like.’
Before a 1:1 with your manager
Paste the meeting agenda or last meeting’s notes and ask for questions that move stalled topics forward instead of just status updates.
Before presenting to a board or panel
Paste your own deck and ask AI to play the audience: what questions would they ask, and where are the weak spots they’d press on.
When it goes wrong
The questions it gives me feel generic
You probably pasted a summary instead of the real document. Paste the full text, including numbers and fine print, so AI has something specific to question.
Some questions feel too pointed to ask out loud
Ask AI to mark which questions are safe to raise in the room and which are better asked privately afterward.
I don’t have a document, just a verbal agenda
Type out the agenda bullets yourself first, even roughly, then run the same prompt on that.
When not to use this
Skip this for casual catch-ups or relationship meetings with no document and no decision at stake. Forcing sharp questions into a conversation that’s meant to just be friendly makes you seem like you’re interrogating someone, not preparing.
What you can now do
You walk into meetings with two questions that show you did the reading, instead of nodding along and hoping no one asks you anything.
Part of Making Life Easier With AI, a weekly walk-through of one small thing AI can take off your plate.
Take this further
Act as a blunt hiring manager who has read ten thousand resumes, not a career coach. I will paste my full resume and the job description I want. Rewrite the whole resume for that role, section by section, in this order: summary, experience, skills, education. Rules: every experience line leads with impact, not duty. Use bracketed placeholders like [8 percent] for any number I did not give you, and list at the end every placeholder I need to replace with a real figure. Keep it to one page of text. Plain formatting only, no tables or columns, so screening software can parse it. After the rewrite, tell me the three weakest claims that need evidence before I send this anywhere. My resume: [paste resume]. The role: [paste job description].
Takeaways
- Copy the pre read into your AI assistant the moment it arrives, before the rush right before the call sets in
- Save one fixed prompt asking for 3 clarifying, 2 forward thinking, and 1 risk question so you reuse it instead of improvising each time
- Narrow the AI’s output down to two questions and open with the risk or gap question to set a serious tone early
- Hold onto the flagged gaps, missing owners, vague numbers, unclear dates, as a head start on your follow up email
- Reuse the same prompt structure before interviews, 1:1s, and board presentations, swapping in a job description, agenda, or your own deck as the source document
Questions people actually ask
How do I get AI to give me good questions to ask before a meeting?
Copy the full pre-read into your AI assistant as soon as it arrives, not a summary. Then ask for 3 clarifying questions, 2 questions that show forward thinking, and 1 that surfaces a risk or gap, plus anything vague, missing a number, or missing a named owner. Save that exact prompt so you can reuse it every time.
How many questions should I actually bring into the meeting?
Two. AI will typically return six or seven options, but walking in with a long list means you never get through it. Pick the sharpest clarifying question and the risk question, keep the rest as backup, and lead with the risk question if the meeting format allows it.
What do I do if I only have a verbal agenda, not a document?
Type out the agenda bullets yourself, even roughly, before the meeting starts. Once you have that rough text, run the same 3 clarifying, 2 forward looking, 1 risk prompt on it. The method depends on having specific text to question, not on the text being polished or official.
Can this same method work for a job interview or a 1:1 with my manager?
Yes. For an interview, paste the job description and the company’s about page and ask for questions that show you understand the role. For a 1:1, paste the agenda or last meeting’s notes and ask for questions that push stalled topics forward instead of generic status check ins.
When should I skip this and not force sharp questions into a conversation?
Skip it for casual catch ups or relationship focused meetings where no document and no decision is on the table. Bringing interrogation style questions into a conversation meant to be friendly tends to make you seem suspicious of the other person rather than prepared.
Sources
- The Surprising Power of QuestionsHarvard Business Review
- Meeting Statistics 2026: 100 Data Points on Time, Cost, ProductivityFlowtrace
- Monitoring AI Adoption in the US EconomyFederal Reserve
What happens next
As generative AI use at work keeps climbing past 4 in 10 US workers according to Federal Reserve tracking, simple reusable prep prompts like this one are likely to become a default habit rather than a novelty for people walking into meetings, interviews, and reviews. Watch for calendar and meeting tools to start offering this kind of document based question generation automatically, which would remove the manual copy and paste step this article relies on.
