September 21, 2026
What Is the Kano Model? Classify Features With Customer Feedback
The Kano model sorts product features into must-have, performance and delight categories using a paired survey. Learn the questions, the scoring table and the limits.
TL;DR
Kano classifies every feature by how it moves satisfaction: must-have, performance, delight, indifferent or reverse. Ask each customer two paired questions, one for when the feature is present and one for when it is missing, then map the answers onto the standard evaluation table. The output tells you what to protect, what to improve and what would be wasted effort.
Author: Inputly
What is the Kano model?
The Kano model classifies product features by the way they affect customer satisfaction instead of by how loudly customers ask for them. Every feature falls into one of five categories: must-have, performance, delight, indifferent or reverse. A must-have, such as a working login, earns no praise when present and causes anger when missing. A performance feature, such as upload speed, moves satisfaction in proportion to how good it is. A delight feature, such as a small automation nobody expected, produces satisfaction out of proportion to its cost. The model comes from Noriaki Kano's work on quality attributes, and it survives because it explains a pattern most product teams feel but cannot name: customers rarely tell you which features they take for granted.
TL;DR: Kano sorts features into must-have, performance, delight, indifferent and reverse using a paired survey. Ask each customer one question for the feature being present and one for it being absent, then map the pair onto the evaluation table. Protect the must-haves, compete on the performance features, and spend your bets on delight.
What are the five Kano categories?
| Category | If the feature is missing | If the feature is present | What to do | |---|---|---|---| | Must-have | Strong dissatisfaction | Neutral | Never break it | | Performance | Dissatisfaction | Satisfaction, in proportion | Improve until it stops paying | | Delight | Neutral | Strong satisfaction | Bet here, then expect it to decay | | Indifferent | Neutral | Neutral | Cut it | | Reverse | Satisfaction | Dissatisfaction | Remove or make optional |
The categories are not labels you assign in a meeting. They are the output of a survey, and the survey is short enough to run next to the questions you already ask in a customer feedback survey.
How do you run a Kano survey?
- Pick five to ten candidate features. More than ten makes the survey long, and length is what kills response rate.
- For each feature write two questions: the functional one, "How do you feel if the product has this?", and the dysfunctional one, "How do you feel if the product does not have this?"
- Use one five-point scale for both: I like it, I expect it, I am neutral, I can tolerate it, I dislike it.
- Map each pair of answers onto the evaluation table: functional "I like it" plus dysfunctional "I expect it" is a must-have, while functional "I like it" plus dysfunctional "I am neutral" is a delight.
- Count the categories per feature and take the majority, then segment by plan or by customer type before you conclude anything.
A feature can be a delight for a small agency and a must-have for an enterprise account. That split is usually the single most valuable finding of the whole exercise.
How do you use the results without over-trusting them?
Treat Kano as a classification with an expiry date. A delight becomes a performance feature as soon as a competitor ships it, and a performance feature becomes a must-have once it is standard. Rerun the survey once or twice a year for the features you are still betting on. Also keep the two signals separate: Kano tells you what a feature does to satisfaction, and a prioritisation model such as how to prioritise feature requests tells you what to build this quarter. Run the two together and you stop over-investing in must-haves that already work while neglecting the cheap surprises that earn loyalty. Where you do not know which features to test at all, start with the raw demand on a feature voting board and let the top requests feed the next Kano round.
Frequently Asked Questions
- Is Kano quantitative or qualitative? It is quantitative in form and qualitative in reading. You get counts per category, but the value is the category, not a decimal score.
- How many features should one survey cover? Five to ten. Each feature costs two questions, and every extra question reduces the response rate you need for the result to mean anything.
- What if a feature lands as indifferent? Cut it or hide it. An indifferent feature still costs maintenance, testing and interface space.
- Can you run Kano internally? Yes. Ask support agents and sales teams the same paired questions, then compare the internal view with what customers actually answered.
- What do you do after the classification? Protect the must-haves, improve the performance features that still move the needle, and spend your next bet on one delight, then measure the effect in your feedback analytics.
Run a Kano round without a research budget. Create a free space on Inputly, send the paired questions, and let the categories decide what you build next.