AI can’t read minds, but it can read your lesson and say “When students finish this, they’re going to wonder…” It’s not perfect, but it’s useful.
How to Predict Student Questions
Take the content of one module—your lesson, worksheet, video transcript—and ask Claude: “After a student completes this module on [topic], what are the five most likely follow-up questions they’ll have? Why would each one come up naturally?” Claude will generate realistic questions based on what the lesson covers and what’s adjacent to it.
This works best if your lesson teaches a concept and stops. A lesson on “How to Price Your Offering” will predictably generate: “What if my competitor is cheaper?” “How do I explain my price to skeptical customers?” “Should I ever discount?” These aren’t mystical questions—they’re logical next thoughts.
Pre-Answer the Common Ones
Once you have the predicted questions, ask yourself: Can I answer this in the module itself? “How to Price” should probably include a “Handling Objections” section. If the answer doesn’t fit in the module, pre-write an FAQ or support resource for it. When students ask, you have an answer ready instead of scrambling to write it during the Q&A.
The more you use this with past cohorts, the better your predictions get. Your first prediction might catch 60% of actual questions. By the third cohort, you’ll have refined the module to address most follow-ups inside it, and only unexpected questions remain.
What This Means for Educators
Teaching live means managing live questions. If you can predict and pre-answer 70% of them, your live sessions run smoother. You’re prepared. You don’t have to think on your feet about every question. You can focus on coaching and depth instead of basic clarification.
Try This on One Module This Week
Pick your most-questioned module. Paste the content into Claude with the question-prediction prompt. See how many predicted questions match the real ones you’ve gotten from past students. You’ll be surprised how accurate it is.