KANO model analysis
SurveyMars now supports KANO model analysis for survey research. The KANO model is used to analyze user preferences and ranking of various requirements. It has wide applications in enterprise product requirement research and market studies.
What is KANO Model
The KANO model analyzes user attitudes toward various functional requirements and helps prioritize product features based on their impact on user satisfaction. KANO model data is typically collected through questionnaires with strict format requirements.
For each feature/service, the KANO model uses two questions:
- Positive Question: User's attitude toward a product/service that HAS this feature
- Negative Question: User's attitude toward a product/service that DOES NOT HAVE this feature
Each question has five response options: A) Dislike, B) Can tolerate, C) Neutral, D) Expected, E) Like. These are scored as 1, 2, 3, 4, and 5 respectively, where 1 represents "Dislike" and 5 represents "Like".
KANO Attribute Categories
The KANO model classifies features/services into six attribute categories:
| Attribute | Characteristics |
|---|---|
| Attractive (A) | Features that exceed user expectations. High improvement in this feature significantly increases satisfaction, but absence does not significantly decrease satisfaction. |
| One-dimensional (O) | Features that increase satisfaction when present and decrease satisfaction when absent. |
| Must-be (M) | Features that do not increase satisfaction when present, but significantly decrease satisfaction when absent. |
| Indifferent (I) | Features that do not affect satisfaction whether present or absent. |
| Reverse (R) | Features that increase satisfaction when absent. |
| Questionable (Q) | Results indicating users did not understand the question or answered incorrectly. |
Priority Order: Must-be attributes > One-dimensional attributes > Attractive attributes > Indifferent attributes. Reverse attributes should not be provided.
Creating KANO Model Questions
Important: KANO model questionnaire design has strict format requirements. You must follow this format exactly.
1. For each feature/service you want to analyze, create two questions:
- Positive Question: "What is your attitude toward a product that HAS '[Feature Name]'?"
- Negative Question: "What is your attitude toward a product that DOES NOT HAVE '[Feature Name]'?"
2. Each question must have exactly five options:
- A) Dislike (Score: 1)
- B) Can tolerate (Score: 2)
- C) Neutral (Score: 3)
- D) Expected (Score: 4)
- E) Like (Score: 5)
3. Ensure that higher scores represent higher approval, regardless of whether it's a positive or negative question.
4. Repeat this process for all features/services you want to analyze.
5. Share your survey and collect responses.
KANO Model Analysis
1. Navigate to the analysis section of your survey.
2. Click "Analysis Results" and select "KANO Report" from the analysis options.
Understanding KANO Mapping
The system automatically maps each feature/service to one of the six KANO attributes based on the combination of positive and negative question responses. The mapping relationship is shown in the following table:
For example:
- If positive question = "Like" (5) and negative question = "Neutral" (3), the feature maps to "Indifferent (I)" attribute
- If positive question = "Like" (5) and negative question = "Dislike" (1), the feature maps to "One-dimensional (O)" attribute
- If positive question = "Neutral" (3) and negative question = "Dislike" (1), the feature maps to "Must-be (M)" attribute
Understanding KANO Results
The system outputs several types of results:
1. Attribute Distribution Table: Shows the percentage distribution of each feature/service across the six KANO attributes.
2. Feature Classification: Each feature/service is classified based on the attribute with the highest percentage.
3. Better-Worse Coefficient: Calculates two key metrics:
- Better (Satisfaction Impact): = (A+O) / (A+O+M+I), ranges from 0 to 1. Higher values indicate greater sensitivity and higher priority.
- Worse (Dissatisfaction Impact): = -1 × (O+M) / (A+O+M+I), ranges from -1 to 0. Lower values (more negative) indicate greater sensitivity and higher priority.
4. Better-Worse Coefficient Chart: A quadrant chart showing the relationship between Better and Worse values for all features/services.
Interpreting Better-Worse Chart
The Better-Worse coefficient chart is divided into four quadrants:
Quadrant 1 (Top Right): One-dimensional attributes - High Better value, high absolute Worse value. Features in this quadrant should be prioritized.
Quadrant 2 (Top Left): Attractive attributes - High Better value, low absolute Worse value. Features in this quadrant should be prioritized and improved as much as possible.
Quadrant 3 (Bottom Left): Indifferent attributes - Low Better value, low absolute Worse value. Features in this quadrant are typically not provided.
Quadrant 4 (Bottom Right): Must-be attributes - Low Better value, high absolute Worse value. Features in this quadrant must be satisfied.
Case Study Example
Background: A mobile phone feature/service requirement research identified 10 features/services through brainstorming: Projection function, Left/Right hand mode, Super fast charging, Remove SIM card, 3D projection, Photo search, Auto beauty, Anti-theft lock, Remote control, Hand warmer, Telescope, Microscope. A total of 100 valid responses were collected. The goal is to analyze user attitudes toward these 10 features/services using the KANO model to provide recommendations for product development.
Questionnaire Design
For each feature, two questions were designed:
Example for "Projection function":
- Positive Question: "What is your attitude toward a mobile phone that HAS 'Projection function'?"
- Negative Question: "What is your attitude toward a mobile phone that DOES NOT HAVE 'Projection function'?"
Each question had five options: A) Dislike (1), B) Can tolerate (2), C) Neutral (3), D) Expected (4), E) Like (5)
Results Summary
Based on the attribute with the highest percentage, the 10 features were classified as follows:
| Attribute | Features | Count |
|---|---|---|
| One-dimensional (O) | Projection function | 1 |
| Attractive (A) | Left/Right hand mode, Super fast charging, Photo search | 3 |
| Must-be (M) | Auto beauty, Anti-theft lock | 2 |
| Indifferent (I) | Remove SIM card, 3D projection, Remote control, Hand warmer, Telescope, Microscope | 6 |
Better-Worse Analysis
Quadrant Analysis:
- Quadrant 1 (One-dimensional): Projection function - Should be prioritized
- Quadrant 2 (Attractive): Super fast charging, Left/Right hand mode, Photo search - Should be prioritized and improved as much as possible
- Quadrant 3 (Indifferent): Other features - Typically not provided
- Quadrant 4 (Must-be): Auto beauty, Anti-theft lock - Must be satisfied
Product Development Recommendations
Based on the priority order (Must-be > One-dimensional > Attractive > Indifferent), the mobile phone manufacturer should:
1. First develop Must-be attributes: Auto beauty and Anti-theft lock (2 features)
2. Urgently develop One-dimensional attributes: Projection function
3. Develop Attractive attributes: Left/Right hand mode, Super fast charging, and Photo search (3 features) - improve as much as possible
4. Ignore Indifferent attributes: The remaining 6 features can be ignored as they do not significantly affect user satisfaction
FAQ
Q1: Why must data be scored from 1 to 5?
A: The KANO model requires exactly five response options for each question (both positive and negative). Therefore, only scores 1, 2, 3, 4, and 5 can be used to represent the five options.
Q2: What is the correct data format for KANO model?
A: For KANO model analysis, higher scores must represent higher approval, regardless of whether it's a positive or negative question. Both positive and negative questions should follow this rule.
Q3: Why do the attribute table and Better-Worse chart sometimes show inconsistent results?
A: The KANO model can use both the attribute distribution table and the Better-Worse chart to determine feature attributes. While both serve the same purpose, they calculate from different angles, which may occasionally result in inconsistent attribute classifications. This is normal.
Q4: How can I modify chart labels in KANO model output?
A: The system automatically generates quadrant charts based on feature names.
Q5: What is the priority order for KANO attributes?
A: The priority order is: Must-be attributes > One-dimensional attributes > Attractive attributes > Indifferent attributes. Reverse attributes should not be provided. This order ensures that essential features are developed first, followed by features that enhance satisfaction.
Important Notes
- KANO model questionnaire design has strict format requirements. You must have both positive and negative questions for each feature/service, and scores must be properly labeled (1=Dislike, 2=Can tolerate, 3=Neutral, 4=Expected, 5=Like).
- Data must be scored from 1 to 5 only.
- For both positive and negative questions, higher scores must represent higher approval. If your data does not follow this format, use data encoding to convert the numbers.
- The KANO model classifies features into six attributes. Final priority should be determined based on attribute characteristics, using both the attribute distribution table and Better-Worse coefficient chart.
- Product development priority order: Must-be attributes > One-dimensional attributes > Attractive attributes > Indifferent attributes. Reverse attributes should not be provided.
- The Better-Worse coefficient chart provides a visual comparison of all features/services. Higher Better values and more negative Worse values indicate higher priority.
