In 1984 Noriaki Kano, a professor in Tokyo, published a way of thinking about product features that's still useful today. His insight was simple: not every feature affects satisfaction the same way. Some only matter when they're missing. Some make people happier the more you have of them. Some delight people precisely because nobody expected them.
You don't need the full survey method to use the idea. You need the three categories and an honest afternoon.
The three that matter most
Must-haves. Nobody thanks you for these, and everybody leaves without them. Signing in that works. Data that doesn't vanish. An export when you need to leave. If a must-have is broken, nothing else you ship will be noticed.
Performance features. More is better, in a straight line. Faster search, more integrations, longer history. Customers can tell you about these directly, and they show up on feedback boards constantly.
Delighters. Things people didn't know to ask for and love when they find them. They rarely appear as feature requests, because nobody requests what they can't imagine.
Kano also named features people don't care about either way, and features some people actively dislike. Both are worth spotting before you build them.
Read your board through the three
Go through the top thirty posts and put each in a bucket. You'll probably find:
- A handful of must-haves in disguise. "Export to CSV" often isn't a nice-to-have. It's someone telling you they can't do their job. These deserve to jump the queue.
- A long list of performance asks. This is where votes are most useful, because more people wanting more of something is exactly what they measure.
- Very few delighters. That's normal. Your board won't tell you about them. Watching people use the product will.
Things drift down
Kano noticed something else: features slide over time. Today's delighter is next year's performance feature and the year after's must-have. Dark mode was a delight once. Now people file it as a bug when it's missing.
So revisit the buckets once or twice a year. A request you parked as "nice to have" may have quietly become something people expect.
A shortcut for the survey
The full method asks two questions about each feature: how would you feel if it were there, and how would you feel if it weren't. You can borrow that on a smaller scale. When a request is near the top of your list, ask a few of the people who voted both questions in a short message. The pattern of answers tells you which bucket it's really in.
Kano isn't science, but it starts a better conversation than "how much do you want this?", and that's usually enough.