AI Analysis for PMs: From Feedback Chaos to Decisions in Hours
Caitlin Sullivan
Trains teams at YouTube, Canva, Maze +
Build AI analysis systems you can actually bet your roadmap on.
Most PMs think their AI analysis problem is prompting.
It's not. It's thinking.
AI doesn't analyze feedback. It mirrors how YOU would analysis feedback - faster, and more confidently.
If your analysis instincts are shallow, AI scales that - and makes mistakes bigger, too.
This is why most AI-generated insights sound smart but say nothing.
"Users want {feature}" can be straight up wrong - even when AI makes it look like there's evidence - unless you know how to push it.
〰️
This course does two things for PMs:
It rewires how you think about customer evidence — signals that matter, contradictions worth chasing.
Then it gives you AI systems that operationalize that thinking — so you get defensible insights in hours, not summaries you secretly don't trust.
〰️
🚀 Bring your own project + tough decisions.
You'll finish every week with real insights from your actual data — not practice exercises. Work that moves current decisions forward.
〰️
What you'll leave with:
→ Reusable AI workflows for verified insights
→ Prompts that catch hidden signals generic prompting misses
→ Repeatable systems that always deliver
What you’ll learn
You've used AI for feedback before. This is how you actually make it work
Analyze 30+ interviews and 100s of survey responses
Master exact prompting techniques that compress week-long analysis into one focused afternoon
Get to prioritized actions based on traceable evidence, not "users want better onboarding"
Learn coding techniques that actually reveal specific patterns, not generic "improve UX" themes
Force accurate calculations so your numbers are bulletproof
Find the signal hiding in the noise that competitors miss, (because their AI processes are weak)
Implement processes for checking outputs, without understanding Python
Build your first analysis agent that runs while you focus on big decisions
Set up continuous feedback loops that flag changes automatically
Use bulletproof validation steps that catch hallucinations before they embarrass you
Build quality checks that take minutes, not hours
Learn directly from Caitlin
Caitlin Sullivan
Ex-Head of UXR in Product Team - Spotify Business, Advisor, Top 100
Who this course is for
You're a PM expected to make evidence-based roadmap decisions on solid discovery -- but you often lack time to be as thorough as you'd like.
You've spent hours digging through feedback to find signal - or cleaning up after AI's bad job at analysis tasks. You need speed + rigor.
You're a CPO / Product leader who wants your team to use reliable AI tactics for decisions based on evidence. Your ROI: Level everyone up.
Prerequisites
AI Adoption
You've used ChatGPT, Claude, or other LLMs. Ideally, your company has an approved LLM so you can apply lessons with your data immediately.
Your Role
You're a PM or similar, responsible for making product decisions based on customer feedback.
Testing with Data
You have access to customer feedback you want to analyze —or are willing practice with my data.
What's included
Live sessions
Learn directly from Caitlin Sullivan in a real-time, interactive format.
Practice tactics together LIVE to get from data to insights every week
Each session takes you from raw data to finished output for one specific analysis task. Hands-on work with your own data or mine, plus Q&A to get unstuck.
Lifetime access
Go back to course content and recordings whenever you need to. Previous course participants refer to my content as their "continuous playbook" that serves them in use cases down the line that they hadn't even thought of during the course.
Learn from real PM problems, not just your own
Group Q&A during live calls means you'll see how others tackle similar data messes and steal their solutions. This has been one of the hidden gems of previous course cohorts!
Certificate of completion
Share your new skills with your employer or on LinkedIn.
UPDATED content for 6 months
AI moves fast; you'll get new tactics from future cohorts automatically
Maven Guarantee
This course is backed by the Maven Guarantee. Students are eligible for a full refund up until the halfway point of the course.
Course syllabus
3 live sessions • 37 lessons • 3 projects
Week 1
Learn what "good" looks like: LLMs + Solid Analysis
Analyzing interviews like an expert with AI
Feb
4
Transcripts to Insights in ≈1hr + AMA (Prompting + Interview Analysis foundations)
Week 2
200 NPS comments → the 3 things that actually matter
From Survey Data to Product Insights
Feb
11
100's of Short Customer Feedback Responses -> Priorities (in < 1.5 hours)
3 sources, 1 story: synthesis that holds up
Free resources
Schedule
Live sessions
1-2 hrs / week
Each session takes you from raw data to finished output for one specific analysis task. You'll work hands-on during the session—using your own customer data or datasets I provide—so you leave with real progress, not theories and notes. The last 20 minutes are open Q&A to get you unstuck on your specific situation.
Wed, Feb 4
4:30 PM—6:00 PM (UTC)
Wed, Feb 11
4:30 PM—6:00 PM (UTC)
Wed, Feb 18
4:30 PM—6:00 PM (UTC)
Projects (optional)
1 hr / week
Each week has one project designed to make you apply what we covered to your actual work. Bring your own feedback data, or use data I provide to complete the practice.
Async content
2 hrs / week
This course is packed with video walkthroughs, explainers, worksheets and frameworks you can go through on your own schedule around each session. I flag videos to watch before our weekly live practice sessions, and everything else is flexible.
Testimonials
I gained confidence on applying AI, as well as ideas for new areas where AI could be applied.
I've struggled with some sort of mental bias or imposter syndrome when it comes to heavy AI use, and this course alleviated that (seeing a distinguished professional using AI with great results).
Ilkka
Senior Product Manager at SupermetricsA valuable, insightful and practical course that supercharges the value of research when we (an experienced human) partner with AI. I loved Caitlin’s style - the perfect balance of principles, examples and plenty of prompts 😀 Highly recommended for anyone working in or leading a product team. This course will show you the art of the possible!
Nick
AI Product Coach, SuperHuman Teams LtdThe course is full of insights on using different LLMs, conducting thorough analysis and learning insights that aren't easy to spot during manual review, using different approaches to prompting and digging deep into research findings, creating own systems for analyses and so much more. 🙌
Marija
Customer Success at DoistThis course is an amazing investment to immediately improve the quality and speed of AI assisted research analysis across LLMs. Reliable, rigorous principles and tactics that I'll return to daily.
Michelle
Staff Experience Researcher at RampThis course helped me unlock more functionality from AI than I believed existed...I am now able to do qual analysis much faster and I have higher personal confidence in the results.
Anna
UX Researcher at MetaThis course completely changed how I use AI in research. I learned to turn my “in my head” synthesis into clear, step-by-step prompts, and use AI to speed up transcript analysis. It helped me cut repetitive work and get sharper insights from customer interviews much faster.
Prun
Product Designer
Frequently asked questions
€599
EUR