Demand Analysis — KANO Model

Background

As a product manager, John often has difficulty in meeting product demands, because developers are pre-occupied while users have various demands. How can John define the real demands of users and give priority to important demands with limited development resources?

John decided to introduce KANO model to systematically define, analyze, and refine the demands to improve efficiency.

Analysis Process

KANO model is a useful tool for classifying and prioritizing user demands, based on analyzing the impact of demands on user satisfaction, which shows the non-linear relationship between product performance and user satisfaction.

John surveyed about 100 users and drew a four-quadrant diagram with the KANO model.

The four quadrants correspond to four types of demands, and their priority order is: must-be demands > one-dimensional demands > attractive demands > indifferent demands.

• Must-be demands (must have): Pain points. For users, these demands must be met and are taken for granted. If must-be demands are not met, user satisfaction will decrease significantly. These are the core demands that must be considered when designing a product.

• One-dimensional demands (should have): If these demands are met, user satisfaction will increase; otherwise, user satisfaction will decrease. One-dimensional demands are significant in comparing competitive products.

• Attractive demands (may have): Functions that delight users, which are beyond users' expectations and increase users' loyalty. If attractive demands are not met, user satisfaction will not decrease.

• Indifferent demands (optional): Demands that users do not care about and do not affect user experience. Avoid designing functions of this type.

Analysis Result

John shared this dashboard with colleagues and decided to add Function 2, Function 3, Function 5, and Function 8. With data analysis, John's proposal is accepted and there is no argument about which functions to add, improving the efficiency.

Getting the Dashboard

Click to preview the online dashboard: KANO Model. You can save it, view the editing steps, and practice.

Designing a Questionnaire

In a KANO questionnaire, every function should have one positive question and one negative question.

After the survey, you need to clean the data. Click to download the cleaned data in this example: KANO Raw Data.xlsx

Processing Data

1. Create an analysis subject named KANO Analysis, add KANO Raw Data, click Field Settings, and tick all fields.

2. Add a Formula Column named Merged Attitudes to merge the columns Attitudes towards adding function and Attitudes towards not adding function.

Finally, identify the linkage between functions and the demands of users according to Attitudes towards adding function and Attitudes towards not adding function.

M: Must-be demand; O: One-dimensional demand; A: Attractive demand;

I: Indifferent demand; R: Reverse demand; Q: Questionable result.

3. In the last step, you have defined the types of demands. Now you should add the column Type to the table.

You can copy the formula and replace Merged Attitudes with the field name you created. This formula uses the switch function.

SWITCH(Merged Attitudes,"Like it very much Like it very much","Q","Like it very much As it should be","A","Like it very much Indifferent","A","Like it very much Reluctantly accepted","A","Like it very much Dislike it very much","O","As it should be Like it very much","R","As it should be As it should be","I","As it should be Indifferent","I","As it should be Reluctantly accepted","I","As it should be Dislike it very much","M","Indifferent Like it very much","R","Indifferent As it should be","I","Indifferent Indifferent","I","Indifferent Reluctantly accepted","I","Indifferent Dislike it very much","M","Reluctantly accepted Like it very much","R","Reluctantly accepted As it should be","I","Reluctantly accepted Indifferent","I","Reluctantly accepted Reluctantly accepted","I","Reluctantly accepted Dislike it very much","M","Dislike it very much Like it very much","R","Dislike it very much As it should be","R","Dislike it very much Indifferent","R","Dislike it very much Reluctantly accepted","R","Dislike it very much Dislike it very much","Q")

4. Add Group Summary to get the number of participants of every function and every type.

For example, 48 participants think Function 1 as Indifferent.

5. The number of participants of every function may differ because some participants may skip questions. Add Group Summary named Number of Participants to calculate the number of participants of every function.

6. Calculate the proportion of demand type of every function.

For example, the proportion of Type I of Function 1 is 0.48.

Creating a Component

1. Copy five Proportion fields.

2. Set Detail Filtering for the copied field Proportion 1. The condition is Type In Fixed Value A. Rename the field Proportion A.

Repeat the above steps to set the other copied fields.

3. Use Better-Worse indicators.

• Better is the satisfaction indicator of adding a function. Better = (Proportion A + Proportion O) / (Proportion A + Proportion O + Proportion M + Proportion I). The closer Better is towards 1, the better the improvement of user satisfaction will be and the faster it will increase.

• Worse is the dissatisfaction indicator of not adding a function. Worse=-1*(Proportion O + Proportion M) / (Proportion A + Proportion O + Proportion M + Proportion I). The closer Worse is towards 1, the greater impact it has on user dissatisfaction and the faster the dissatisfaction decreases.

Add calculation indicators according to the above two formulae named Better and Absolute Value of Worse (You do not need to multiply the absolute value of Worse and -1.).

4. Select Scatter Chart, and drag Better and Absolute Value of Worse. Drag the Function field into Color and Label in Graphic Properties.

5. Add a vertical cordon of the average value of Better and a landscape cordon of the average value of Absolute Value of Worse.

Demonstration

The effect is shown in section "Analysis Process".