Presentation Feedback AI Prompts for Speakers

Most speakers improve slowly because feedback is slow. You give a presentation, wait days or weeks for evaluation forms, receive generic comments (“great job!”), and then prepare your next presentation without clear guidance on what to change. The feedback loop is too slow and too vague to drive improvement.

The alternative is not waiting for others. You can use AI to analyze your presentation delivery, generate specific feedback, and create a continuous improvement practice that does not depend on external evaluators.

AI Unpacker provides prompts designed to help speakers at all levels get actionable feedback on their presentations, from pitch decks to keynotes.

TL;DR

• Traditional feedback loops are too slow and too vague for meaningful improvement.
• AI can provide immediate, specific feedback on content, structure, and delivery.
• Video recording + AI analysis is more effective than self-reflection alone.
• The most improvement comes from focused practice on specific weaknesses.
• Audience analysis is the foundation of every effective presentation.
• Practice sessions should simulate real conditions as closely as possible.

Introduction

Public speaking is a skill that improves with practice and feedback. The problem is that practice without feedback is like golfing with your eyes closed — you might get better by accident, but structured improvement requires seeing what you are doing wrong.

Traditional feedback comes from coaches, colleagues, or audience evaluations. It is valuable, but it is also sporadic, subjective, and delayed. By the time you receive feedback, the presentation is over, your memory of it has faded, and the opportunity for immediate practice has passed.

AI changes the feedback equation. You can record your presentation, transcribe it, and receive detailed analysis within minutes. The feedback is immediate, consistent, and focused on specific improvement areas.

1. Presentation Preparation

The best presentations are built for specific audiences with specific goals. Before you start building slides, you need to understand who you are speaking to and what success looks like.

Prompt for Audience Analysis

Conduct a comprehensive audience analysis for an upcoming presentation.

Presentation context:
- Occasion: Board meeting (quarterly update)
- Audience: 8 board members (mix of investors and industry experts)
- Your role: CEO presenting company performance and strategy
- Duration: 25 minutes + 15 minutes Q&A
- Stakes: This presentation influences board sentiment toward Series B fundraising

What I know about the board:
- 3 members are financially focused (want metrics, ROI, timeline)
- 2 members are industry veterans (want to see competitive positioning)
- 1 member is a former competitor (knows the market well)
- 2 members are investors with limited operational experience

What the board has seen before:
- Previous quarters showed declining growth metrics
- Latest product launch had delays
- Customer churn increased slightly

What I want them to feel:
- Confident in leadership's grip on the business
- Excited about strategic initiatives
- Convinced that the Series B is necessary for acceleration

What I want them to decide:
- Approval of revised OKRs for next quarter
- Go-ahead for Series B preparation

Analysis requirements:
1. What are the board's likely concerns based on their backgrounds?
2. What messages will resonate with each board member type?
3. What objections might arise and how should I address them?
4. How should I structure the presentation to build toward the decision?

Tasks:
1. Map audience motivations (what does each group care about most?)
2. Identify potential friction points (what might cause skepticism?)
3. Recommend content prioritization (what to emphasize vs. minimize?)
4. Suggest Q&A preparation strategies

Generate an audience analysis report with specific recommendations for this presentation.

2. Delivery Self-Assessment

Before you ask AI to analyze your presentation, you need to have something to analyze. Recording yourself is the first step. But even before that, structured self-assessment helps you identify the areas where you most need feedback.

Prompt for Self-Assessment Framework

Create a self-assessment framework for my presentation delivery.

Presentation type: Conference keynote (public speaking, 40-minute session)
Audience: 400 attendees (mix of practitioners and executives in my industry)
Experience level: Intermediate speaker (presented at company all-hands, smaller workshops, but never a 400-person keynote)

Areas where I know I struggle:
- I speak too fast when nervous
- I have a habit of saying "um" and "uh" frequently
- I tend to read my slides rather than speak to the audience
- I do not know what to do with my hands

Areas where I think I am strong:
- I am good at explaining complex concepts simply
- I use analogies effectively
- I connect with individuals in small groups

What I want from AI feedback:
- Identify specific verbal fillers (not just "you say um too much")
- Analyze my pacing across different parts of the talk
- Evaluate whether my slides add value or distract
- Assess whether my transitions work

Self-assessment requirements:
1. Record and review my practice sessions:
   - What specific things to look for in each recording?
   - How to overcome review anxiety that makes me avoid listening?

2. Create a delivery scoring rubric:
   - What dimensions to rate myself on?
   - What does "good" look like vs. "needs work"?
   - How to track improvement over multiple practice sessions?

3. Identify my priority improvement areas:
   - Which weakness, if fixed, would have the biggest impact?
   - How to structure practice to focus on that weakness?

4. Develop pre-presentation routines:
   - What to do in the 30 minutes before speaking?
   - How to manage nerves without medication?

Generate a self-assessment framework with specific rubrics and practice protocols.

3. Post-Presentation Analysis

After a presentation, you have the best data for improvement. But most speakers move on to the next thing without analyzing what happened. AI can help you extract specific insights from your presentation performance.

Prompt for Post-Presentation AI Analysis

Analyze this presentation recording and provide specific feedback.

Presentation details:
- Duration: 22 minutes
- Occasion: Sales kickoff presentation to 50 sales reps
- Content: New product positioning and competitive differentiation
- Audience: Sales team learning new product to sell

What I know subjectively:
- I felt good about the middle section (where I told the customer story)
- I know I rushed the opening (felt nervous at the start)
- I ran 3 minutes over time

Self-assessment scores (pre-analysis):
- Pacing: 6/10 (rushed in some places)
- Clarity: 7/10 (clear overall, some technical jargon)
- Engagement: 6/10 (hard to read the room on video)
- Confidence: 5/10 (visibly nervous at start)

Analysis I want:
1. Verbal delivery analysis:
   - Count of filler words (um, uh, like, so, you know)
   - Speaking pace (words per minute, variation across sections)
   - Volume variation (did I modulate effectively?)

2. Content analysis:
   - How well did the opening hook the audience?
   - Was the core message clear?
   - Did the evidence support the claims?
   - Was the call to action compelling?

3. Slide effectiveness:
   - Did slides support the speaker or compete for attention?
   - Were slides visual (not text-heavy)?
   - Did transitions between slides work verbally?

4. Specific improvement recommendations:
   - Top 3 things to change in the next version
   - One thing to practice before next presentation
   - One thing that was actually good (do not just focus on negatives)

Tasks:
1. Generate specific feedback for each analysis dimension
2. Prioritize recommendations by impact
3. Create a practice plan for addressing top issues

Generate a detailed feedback report with specific, actionable recommendations.

4. Continuous Improvement System

One presentation analysis is not enough. You need a system for continuous improvement that builds on each presentation and tracks your progress over time.

Prompt for Improvement Tracking System

Design a presentation improvement system for ongoing development.

Current situation:
- I give 2-4 presentations per month (mix of internal and external)
- I record most presentations but rarely review them
- I want to improve my public speaking skills over the next 6 months
- I have 2-3 hours per week I can dedicate to presentation practice

Goals (ranked):
1. Reduce filler words by 50% within 3 months
2. Speak with more authority and confidence
3. Improve audience engagement (measured by post-presentation surveys)
4. Build a library of stories/analogies I can reuse

Improvement system requirements:
1. Practice routine:
   - What should my weekly practice schedule look like?
   - How to balance new presentation prep vs. skill development?
   - What specific exercises improve each goal area?

2. Recording and review protocol:
   - How often to record myself?
   - What to look for in each review?
   - How to overcome the discomfort of watching myself?

3. Feedback integration:
   - How to incorporate feedback from colleagues?
   - How to use AI feedback alongside human feedback?
   - How to prioritize conflicting feedback?

4. Progress tracking:
   - What metrics to track over time?
   - How to measure confidence improvement?
   - How to know when I have achieved my goals?

5. Resource allocation:
   - 2-3 hours per week: how to spend it optimally?
   - What external resources (coaches, courses) are worth the investment?

Tasks:
1. Design a 6-month improvement plan with milestones
2. Create weekly practice templates
3. Develop a feedback journal format
4. Build a progress dashboard with specific metrics

Generate a complete presentation improvement system with specific protocols and timelines.

FAQ

How do I overcome the discomfort of watching myself on video?

Start by watching with the sound off. Observe your body language, movement, and visual presence before you tackle the verbal content. Then watch with sound but without analyzing — just notice how it feels. Finally, watch with a specific analysis focus (one thing at a time). The discomfort decreases as you build a practice of review. It never disappears entirely, which is actually healthy — it keeps you honest about your performance.

Should I use AI feedback or human feedback?

Both. AI feedback is consistent, immediate, and specific to delivery patterns. Human feedback provides context about audience experience that AI cannot capture. Use AI for technical aspects (pacing, filler words, slide effectiveness) and humans for strategic aspects (audience connection, emotional impact, credibility).

How long does it take to meaningfully improve?

You will see measurable improvement in 30-60 days with consistent practice. The key is focused practice on specific weaknesses, not general repetition. If you practice the same way without structure, you will plateau. If you identify one weakness and drill it specifically, you will improve faster.

Conclusion

Presentation skills improve through deliberate practice, and deliberate practice requires feedback. The traditional model — wait for evaluation forms, hope for specific comments — is too slow for anyone serious about improvement.

AI Unpacker gives you prompts to create feedback loops that are immediate, specific, and continuous. But the willingness to watch yourself honestly, to practice the uncomfortable parts, to prioritize improvement over comfort — that commitment comes from you.

The goal is not a perfect presentation. The goal is a speaker who is always getting better.