How to avoid the trap of flattering feedback
One of the biggest traps in early-stage product development is flattering feedback. It feels great to hear customers say, "I love your product!" or "This is amazing!". But if you stop there, you're missing the real gold.
The truth is, surface-level feedback rarely leads to breakthrough insights. That's why I've built a simple process to dig deeper. Here's what works for me:
1. Look for Patterns, Not One-Offs
Your ICP (ideal customer profile) is like a magnet, it attracts consistent themes. If feedback feels random or scattered, you might not have nailed your ICP yet. But when the same themes come up again and again, you know you're onto something.
2. Ask the Right Questions
I use two simple but powerful tactics to get beyond the surface:
- The "5 Whys" technique to uncover root causes.
- Asking, "Did anything surprise you?" to reveal insights customers might not think to share.
3. Leverage AI to Scale Feedback Analysis
These days to help speed up my analysis, I use a custom GPT to parse feedback at scale. It identifies Jobs to Be Done, gaps, and risks, giving me a structured starting point to refine or act on. This lets me run more experiments than ever before, with a higher margin for error.
4. Balance Speed with Focus
With tools like AI and rapid prototyping, it's tempting to chase every idea. But the key is staying disciplined. Look for patterns, test hypotheses, and double down on what works.
How do you approach customer feedback? Do you have a favourite question or technique that's helped you uncover game-changing insights? Let's swap notes in the comments.
Related
- I once tried to fire a customerI offered our most demanding customer a refund. She told me we were mission-critical. What high-maintenance customers reveal about where your product matters.
- Sometimes pain isn't quite enoughCustomers tolerate a lot of pain. How to tell if a problem is severe enough to pay for: quantify it, understand the alternatives, and test willingness to pay.
- AI Didn't Fix Discovery. It Made the Problem Harder to Ignore.AI made delivery faster, but discovery is still the bottleneck. Rough prototypes surface what requirements docs hide, if you know how to read the friction.