Home Tools Product managementKano survey analyzer

Kano model analysis for your survey

Paste functional and dysfunctional answers or upload a CSV from Google Forms, Typeform or Excel. The analyzer maps every answer pair through the Kano evaluation table, finds each feature’s category, calculates Better and Worse coefficients and plots the features on a chart.

● Free, no sign-upUpdated: September 16, 2026

Answers: respondent, feature, functional, dysfunctionalOne row is one person’s answers about one feature. A wide sheet from a survey form works too: a pair of columns per feature, such as “Feature (F)” and “Feature (D)”, one row per respondent. Comma, semicolon or tab separated.

Upload CSV

Answer scale: 1 — I like it, 2 — I expect it, 3 — I am neutral, 4 — I can live with it, 5 — I dislike it. Words work too.

If the answers are already classified, enter the number of answers in each category for every feature.

Number of answers in each Kano category per feature
Feature	A	O	M	I	R	Q	Actions

—

Features—

Respondents—

Must-be—

Attractive—

Results: category shares, dominant category and coefficients for each feature
Feature	n	A	O	M	I	R	Q	Category	Better	Worse

A — attractive, O — one-dimensional, M — must-be, I — indifferent, R — reverse, Q — questionable. The category is highlighted when it is significantly ahead of the runner-up (Fong’s test).

Better / Worse charthover a point

clear categorycategory not clear

Formulas. Better = (A + O) / (A + O + M + I); Worse = −(O + M) / (A + O + M + I). R and Q answers are left out of the coefficients.

Everything is calculated in your browser — nothing you enter is sent anywhere.

How to use it

Run the surveyAsk two questions per feature: how would you feel if the product had it, and how would you feel if it did not. Answers use the five-point Kano scale.
Paste the answersCopy the columns from a spreadsheet or upload a CSV. The analyzer detects the layout and recognizes numbers and answer wording.
Read the categoriesFor each feature you get A, O, M, I, R, Q shares, the dominant category and whether it clearly beats the runner-up.
Compare on the chartBetter and Worse show how much a feature raises satisfaction and how much its absence hurts. The top-right corner holds the strongest candidates.

What the Kano model is

The Kano model was introduced by Japanese quality researcher Noriaki Kano and colleagues in 1984. Its core idea: user satisfaction does not grow linearly with features. Some features are noticed only when they are missing; others delight people who never asked for them. The model sorts requirements into five categories:

M — must-be. Their absence causes strong dissatisfaction; their presence is taken for granted. Examples: data export in a B2B product, password reset.
O — one-dimensional (performance). The better they are, the more satisfied users become: load speed, number of integrations, search accuracy.
A — attractive (delighters). Nobody misses them, but their presence delights. Over time attractive features turn into one-dimensional ones, and then into must-be.
I — indifferent. Users do not care whether the feature exists.
R — reverse. The feature annoys people who have it, and they are happier without it — pushy notifications, for instance.

A separate category, Q — questionable, covers contradictory answers: the person likes both having and not having the feature. It shows up when a question is unclear or the respondent is not paying attention.

The Kano questionnaire and evaluation table

Each feature gets a pair of questions with the same answer scale:

Functional: “How would you feel if the product had [feature]?” Dysfunctional: “How would you feel if the product did not have [feature]?” 1 — I like it 2 — I expect it 3 — I am neutral 4 — I can live with it 5 — I dislike it

The pair of answers maps to a category through the evaluation table. The analyzer uses the standard version published by Berger and colleagues in 1993: the row is the functional answer, the column the dysfunctional one.

Kano evaluation table
Functional ↓ / Dysfunctional →	Like	Expect	Neutral	Live with	Dislike
Like	Q	A	A	A	O
Expect	R	I	I	I	M
Neutral	R	I	I	I	M
Live with	R	I	I	I	M
Dislike	R	R	R	R	Q

For example, “like” plus “dislike” gives O, while “neutral” plus “dislike” gives M. For every feature the analyzer counts category shares and takes the most frequent one. Ties follow the M > O > A > I rule: the category with the stronger effect on dissatisfaction wins.

Is the category clear?

If 9 people chose A and 8 chose O, calling the feature attractive is a stretch. Fong’s test (1996) checks whether the top category really differs from the runner-up:

|a − b| ≥ 1.65 · √( (a + b) · (2n − a − b) / 2n ) a, b — answers in the first and second categories, n — all answers

If the condition fails, your sample most likely mixes user groups with different expectations. Split the answers by segment — plan, role, tenure — and analyze each one separately.

Better and Worse coefficients

Category shares are hard to compare across features, so Berger and colleagues proposed two customer satisfaction coefficients:

Better = (A + O) / (A + O + M + I) 0 to 1 Worse = −(O + M) / (A + O + M + I) −1 to 0 Example: A = 13, O = 2, M = 0, I = 4 Better = 15 / 19 = 0.79 Worse = −2 / 19 = −0.11

Better (the satisfaction coefficient) shows how much satisfaction rises if you add the feature. Worse (the dissatisfaction coefficient) shows how much it drops if the feature is missing. R and Q answers are not in the denominator.

The chart plots |Worse| horizontally and Better vertically. Four quadrants split at 0.5 mirror the categories: top-left attractive, top-right one-dimensional, bottom-right must-be, bottom-left indifferent. The farther a point is from the origin, the more the feature shapes how people feel about the product.

How to use the results

Ship must-be features first, but do not over-invest: beyond the basic level they add no satisfaction.
Improve one-dimensional features step by step and measure them — this is where competitors are compared.
Attractive features differentiate you for little investment. Point new users to them with a tooltip or a tour.
Indifferent features are candidates for removal from the backlog. Score what remains by reach and effort with the RICE & ICE calculator.

How to run a Kano survey without the usual mistakes

Describe the benefit, not the implementation. “A report lands in your inbox every Monday” is clearer than “export scheduler”.
Keep it to 15–20 features per survey. Each feature is two questions; a long questionnaire produces careless answers and more Q.
Ask your target users. Newcomers and power users often rate the same feature differently — which is where unclear categories come from.
Collect enough answers. With 20 answers per feature a couple of votes flip the category. The survey sample size calculator tells you how many you need for a given precision.
Add an importance question. A “how important is this, 1 to 9” scale helps separate weak delighters from strong ones.
Repeat the survey. Categories drift: yesterday’s delighter becomes a must-be in a couple of years.

Sources

Kano N., Seraku N., Takahashi F., Tsuji S. Attractive Quality and Must-Be Quality. Journal of the Japanese Society for Quality Control, 14(2), 1984 — the original model.
Berger C., Blauth R., Boger D. et al. Kano’s Methods for Understanding Customer-defined Quality. Center for Quality of Management Journal, 2(4), 1993 — questionnaire, evaluation table, customer satisfaction coefficients.
Fong D. Using the Self-Stated Importance Questionnaire to Interpret Kano Questionnaire Results. Center for Quality of Management Journal, 5(3), 1996 — significance of the category.
Matzler K., Hinterhuber H. H. How to Make Product Development Projects More Successful by Integrating Kano’s Model of Customer Satisfaction into Quality Function Deployment. Technovation, 18(1), 1998.

FAQ

How do you analyze Kano survey results?

For each respondent, map the answer pair (feature present, feature absent) to a category with the evaluation table, count category shares for every feature and take the most frequent one. Then calculate the Better and Worse coefficients. The analyzer does all of this from pasted answers.

What questions does a Kano survey ask?

Two per feature: a functional one (“How would you feel if the product had this?”) and a dysfunctional one (“…if it did not?”). Answers: I like it, I expect it, I am neutral, I can live with it, I dislike it.

How are Better and Worse calculated?

Better = (A + O) / (A + O + M + I) and Worse = −(O + M) / (A + O + M + I), where the letters are answer counts per category. Better shows the potential satisfaction gain, Worse the damage done by the feature’s absence.

What if two categories tie?

Formally, the M > O > A > I rule decides. But a tie or an insignificant lead usually means different user groups expect different things. Split the answers by segment and analyze each separately.

How many respondents does a Kano survey need?

There is no strict rule. A practical guide is at least 30–50 answers per feature within one segment: below that, a couple of votes change the category. You can estimate the precision of a share with a survey sample size calculator.

What do the R and Q categories mean?

R is reverse: the feature annoys people and they prefer not having it. Q is questionable: the person likes both having and not having the feature. A large share of Q signals an unclear question.

What data format can I upload?

A CSV or a paste from a spreadsheet: rows of “respondent, feature, functional, dysfunctional”, rows of “feature, functional, dysfunctional”, or a wide sheet with two columns per feature. Answers can be numbers 1–5 or words.