Mike Lyons shows product researchers how to turn simulated customer feedback into testable hypotheses using AI as an analysis partner. He emphasizes spotting patterns in complaints, translating them into theories tied to business outcomes, and proposing observable evidence. The key is staying critical: ask AI to challenge your assumptions and suggest alternatives before you prioritize which tests to run.
Okay, so you've got your feedback from Technova customers. And yes, I know it's a simulation, but now comes the real work, turning those responses into specific questions to investigate through actual customer research. Now, this is where most people kind of stumble. They read through their feedback and either take it at face value or kind of dismiss it entirely. But both of those approaches miss the point. The values in learning how to extract patterns and form assumptions that you can test with real computer customers. And here's a bit of good news. It's the same good news I've been telling you about this whole course. You don't have to do this analysis alone. AI can be your research partner. Okay? So think about what you're really doing when you analyze feedback. You're. You're looking for patterns, you're connecting dots between different responses, and you're translating customer complaints into testable business theories. Well, AI, it excels at this exact kind of pattern spotting. So instead of manually reading through dozens of responses looking for themes, you can ask AI to identify trends across your feedback. So give it all your responses and ask it something like, what are the three most common underlying needs expressed across these different customer types? But hey, don't stop there. Push AI to help you form better research questions by asking it to connect patterns to business outcomes. You might try prompts like based on these patterns, what theories could explain why Technova customers might reduce their platform engagement? You can also use AI to refine your ideas by asking it to identify what evidence would prove or disprove each theory. Prompt it with. If customers struggle to find features, what specific behaviors would we observe in user analytics or customer interviews? The key is treating AI as a thinking partner, not just a pattern finding tool. Let me give you a pro tip here that applies across all of your work with AI. Pro tip, pro tip. You need to be actively thinking alongside the AI, not just passively consuming its output. The moment you disengage your critical thinking AI becomes expensive rather than productive. So ask it to challenge your assumptions, suggest alternative explanations for the patterns you're seeing, and help you prioritize which ideas are worth testing first. You might try prompts like I think customers are confused by our AI task estimation. What are three different theories that could explain the negative feedback about this feature? Each theory suggests different validation approaches, which helps you design better customer research. You're using AI to help you think more systematically about what patterns might mean and how to test your theories with real customers. When you sit down with these actual customers, you'll have clear theories to explore and specific evidence to look for and focused validation methods to try. And that's the real power here. You're using AI not to just spot patterns in customer feedback, but to help you think more deeply about what those patterns might mean and how to test them with your customers.
