Market research tool
Kano model analysis
How results are checked
Named statistical review pending. Selected automated reference cases are published on the validation page.Largest satisfaction lever
0.67CS+
Live courier tracking: one-dimensional
Accurate delivery-time estimate costs the most when missing (CS− -0.73) · 1 of 6 features without a clear category (Fong's test)
Kano analysis sorts product features by how they drive satisfaction. Each feature is asked twice — how would you feel if the product had it, and if it did not — on a five-point scale from "I like it" to "I dislike it". The Kano evaluation table turns each answer pair into a category: must-be features cause dissatisfaction when missing, one-dimensional ones satisfy more the better they are, attractive ones delight without being expected, and indifferent ones barely matter.
Run a Kano analysis on your data
Open a respondent-level file with a functional and a dysfunctional question per feature — coded 1–5 or as text such as "I like it" — and the tool pairs the columns by name (for example wifi_func / wifi_dysf). Or paste A/O/M/I/R/Q counts per feature from an existing report. Everything runs in this browser tab; nothing is uploaded. "Use sample data" loads a synthetic food-delivery app survey with six features and self-stated importance.
CSV, Excel, SPSS, Stata, SAS, R, Parquet, SQLite or JSON. The file is read in this browser tab and never uploaded. For Triple-S, select the .sss or .xml metadata file and its data file together.
One feature per line: name, then the A, O, M and I counts, and optionally R and Q. Separate with tabs (pasted from a spreadsheet), commas or semicolons.
Paste at least one feature with its counts.
Reading the satisfaction–dissatisfaction plot
Each point is one feature. Its height is the satisfaction coefficient CS+ (how much offering it adds), its distance to the right is the size of the dissatisfaction coefficient |CS−| (how much leaving it out costs). Upper right: one-dimensional. Lower right: must-be — get these right first. Upper left: attractive — differentiators once the basics are in place. Lower left: indifferent.
Categories drift over time: today's attractive feature tends to become one-dimensional and then must-be as competitors copy it. Kano results also differ by segment, so compare groups by crosstabbing the answers in the workspace.
The method
Formula
CS+ = (A + O) / (A + O + M + I) CS− = −(O + M) / (A + O + M + I)
where
- A, O, M, I
- number of respondents classified attractive, one-dimensional, must-be and indifferent for the feature
- R, Q
- reverse and questionable answers: counted in the shares, left out of the coefficients
- CS+
- satisfaction coefficient (also called Better), 0 to 1
- CS−
- dissatisfaction coefficient (also called Worse), −1 to 0
- Fong's test
- the top category is clear when |a − b| ≥ 1.65·√((a + b)(2n − a − b) / 2n), with a and b the two largest counts and n all classified answers
Worked example
In the synthetic sample, 294 respondents answered both questions about live courier tracking. 41.8% were one-dimensional, 21.1% attractive and 16.3% must-be, so CS+ = (62 + 123) / (62 + 123 + 48 + 43) = 0.67 and CS− = -0.62. Fong's test: the gap between one-dimensional and attractive is 61, above the critical 18.6, so the feature is clearly one-dimensional.
An accurate delivery-time estimate is must-be (48.1%, CS− -0.73): nobody praises it, but missing it hurts most. A free birthday dessert is attractive (50.3%, CS+ 0.62, CS− only -0.15). Recipe videos and promotional push notifications are indifferent, and 34.9% of respondents answer the push notifications in reverse — they would rather not have them.
One-tap reorder shows why the test matters: 89 one-dimensional against 85 attractive answers is a gap of 4, well below the critical 18.2. There is no clear category; its CS+ of 0.62 still says it adds satisfaction. Recipe videos (57.5% indifferent) can wait.
Using it well
When to use it
Use it when
- Use: Deciding which features to build, keep or drop in a product or service roadmap.
- Use: Separating basic expectations from differentiators before a pricing or conjoint study.
- Use: Checking how features are perceived across segments or over time.
Not the right tool when
- Avoid: Features respondents cannot picture — explain or show each one, or the answers drift to indifferent.
- Avoid: Small samples: with a few dozen respondents per feature, Fong's test rarely finds a clear category.
- Avoid: Ranking features against a budget — combine Kano with importance ratings, MaxDiff or conjoint.
How to interpret it
Rule of thumb
Fix must-be features first, invest in one-dimensional features in proportion to their CS+ and |CS−|, and use attractive features to stand out. When Fong's test finds no clear category, read the coefficients rather than the mode, and use self-stated importance to break ties. A high reverse share means a segment actively does not want the feature.
Questions
Frequently asked questions
What are the Kano categories?
Attractive (A): delights when present, no harm when absent. One-dimensional (O): more is better, less is worse. Must-be (M): expected; its absence causes dissatisfaction. Indifferent (I): makes little difference either way. Reverse (R): the respondent prefers not to have it. Questionable (Q): contradictory answers, such as liking both having and not having the feature.
How is the Kano evaluation table read?
The functional answer selects the row and the dysfunctional answer the column. 'I like it' if present and 'I dislike it' if absent is one-dimensional; 'I like it' if present and expect, neutral or tolerate if absent is attractive; expect, neutral or tolerate if present and 'I dislike it' if absent is must-be; the middle answers on both are indifferent.
What are CS+ and CS− (Better and Worse)?
Berger and colleagues' customer satisfaction coefficients. CS+ = (A + O) / (A + O + M + I) is the share of respondents whose satisfaction rises when the feature is offered; CS− = −(O + M) / (A + O + M + I) is the share whose satisfaction falls when it is missing. They summarise the whole mix of answers, not only the most frequent category.
What is Fong's test?
A check of whether the most frequent category is really ahead of the runner-up. With a and b the two largest counts and n the number of classified answers, the difference is significant when |a − b| is at least 1.65 × √((a + b)(2n − a − b) / 2n). Otherwise the feature is reported as having no clear category.
What is the difference between discrete and continuous Kano analysis?
Discrete analysis assigns each feature its most frequent category. Continuous analysis keeps the full distribution — here through the CS+ and CS− coefficients and the quadrant plot — which is more informative when respondents split between categories.
Sources and next steps
References & further reading
- [1]Kano, N., Seraku, N., Takahashi, F., & Tsuji, S. (1984). Attractive quality and must-be quality. Journal of the Japanese Society for Quality Control, 14(2).
- [2]Berger, C., Blauth, R., Boger, D., et al. (1993). Kano's methods for understanding customer-defined quality. Center for Quality of Management Journal, 2(4), 3–36.
- [3]Fong, D. (1996). Using the self-stated importance questionnaire to interpret Kano questionnaire results. Center for Quality of Management Journal, 5(3).
- [4]Matzler, K., & Hinterhuber, H. H. (1998). How to make product development projects more successful by integrating Kano's model of customer satisfaction into quality function deployment. Technovation, 18(1), 25–38.
Try it on your own data — free, no signup
Upload a CSV, Excel, SPSS, Stata, SAS, R, Parquet or JSON file. Everything runs in your browser; your file never leaves your device.
Open the workspace →