Skip The Qualitative Research? That's A mistake.

Yes, quantitative data matters, but it tells you what is happening, and you need to know why it's happening, because if you don't understand why, you're optimizing blind. Here's what qualitative research actually is and how to do it well. The worst interviews ask customers what they want, while the best ones ask them to reconstruct a real decision. For example, I might say to a research participant, "Walk me through the last time you bought from this brand. What triggered the need? What alternatives did you consider? What almost stopped you? What finally pushed you over the edge?" That sequence (or decision journey) reveals the actual behavior driving the criteria people use, not the ones they think they use. But wait, there's more!

There's a secondary output most people miss, which is the exact words customers use to describe their pain, and when you hear words like "risky," "complicated," or "a time sink," pay attention because while they are interesting adjectives, they're more importantly mental models that will quietly kill your conversion rate unless you address them directly.

Customers normalize friction because they work around broken things so often that they stop seeing them as broken. They won't mention the workaround in an interview because it doesn't occur to them that it's a problem. Watch people. Don't just ask them. Watch someone attempt onboarding. Watch them compare your product to a competitor. Watch them complete a real workflow. You'll find hidden steps, invisible drop-offs, and assumptions baked into your product that nobody has ever surfaced in a survey. In B2B, this means watching how tools move across teams and how decisions actually get approved, not how the org chart says they do. In consumer categories, it means watching shopping, unboxing, and repeat usage. When you're studying habit formation, subscription usage, or service relationships, ask customers to record moments of value and moments of frustration as they happen. The map you get back is richer and more honest than anything reconstructed in a one-hour call.

There are, of course, two failure modes worth knowing:
1. Over-generalizing from loud voices, where a few strong personalities can make a pattern feel universal when it isn't.
2. Confusing what customers say they want with what they'll actually do. Customers are honest about their feelings, and they're often wrong about causes.

The fix is simple but requires discipline: treat every qualitative finding as a hypothesis, not a conclusion, then go confirm or refute it with behavioral data and experiments.

Qualitative research doesn't replace measurement, but it will tell you what's worth measuring.