Kano Model

The Kano Model was created by Kano Noriaki, a professor of quality management at the Tokyo University of Science in 1984 to understand how users perceive the value of certain product capabilities.

In this article I want to provide a brief overview of the methodology. For a detailed reference and different ways to evaluate the results, I recommend The complete guide to the Kano model.

Why using the Kano Model?

The Kano method helps to answer:

• How do we measure satisfaction with certain features?
• How do we understand how those features are perceived by users?
• What features will deliver a wow effect?
• What features must be included to satisfy the basic expectations

With this method it is easy to group features in four categories based on satisfaction and functionality and visualize it as two dimensional graph. This will support decision making but will not provide an answer why users see it this way. Answering the why-question will require additional research work.

How does it work?

1. Craft a survey with each feature listed separately
2. Participants can select a predefined response if the feature would be present or absent
3. Data is analysed and categorized to support the decision making process

The four Kano categories

Indifferent

Indifferent features are those that users don’t care about. They don’t increase satisfaction when present, and they don’t decrease satisfaction when absent.

Must-be

Must-be or basic expectation features are features that are expected by default. If they are missing, users feel frustrated or even betrayed. If those features are there, users will not compliment you, they simple assume that those features should be there. Identifying must-be features is crucial as this helps to understand the minimum feature set you must invest in.

We expect from a phone that we can easily make a phone call. If the interface of a smart phone makes it hard to make simple calls or the device gets frozen, they will simply return the product.

Performance

Performance are the more the better features. Satisfaction rises as performance improves and drops as performance weakens.

How long the battery of your phone lasts under heavy use or how fast the phone charges has a direct impact on satisfaction. The same is true for Wi-FI range and speed, audio quality, etc.

Attractive

Attractive features (delighters or exiters) are features users don’t expect. So, users will not get upset if those features are not there but when they are there, satisfaction jumps exponentially. They create the wow, I did not know the product can do that or hey, that’s nice feeling.

Imagine 2007, a user switching from any mobile phone to an iPhone 1.0. The additional capabilities and user interface of a smart phone felt mesmerizing – wow. Today, being able to interact with your phone in a convenient way, take pictures, browse the internet are elementary features nobody would market anymore. They are simply expected and not having those features would immediately create frustration.

Same holds true for 2001, where WiFi in hotel rooms was still news. Even hot water in a hotel room has been a wow factor at some time. The important aspect is, that attractive features turn into must-be or performance features over time.

Kano survey

Functional / Dysfunctional question

You can either show (mockups), offer a functional prototype or describe 15–20 features.

For each feature ask

• Functional aspect: If you had (feature), how do you feel? (positive or functional question)
• Dysfunctional aspect: If you didn’t have (feature), how do you feel? (negative or dysfunctional question).

The participant must provide the Kano rating for each question:

• like: I like it
• expect: I expect it / It should be that way
• neutral: I’m neutral / I don’t care
• tolerate: I can tolerate it / I can live with it
• dislike: I dislike it

Include a test-run example with explanation at the beginning to allow participants to familiarize themselves with the tool.

Prefer experience or showing the features over describing the features. Let us assume, you want to find out the satisfaction of a certain file upload. The closer it will be to the real experience, the better users will be able to provide feedback.

Self-stated importance question

Often the Kano questions can be optionally extended with a self-stated importance scale from 1 (least important) to 9 (most important) to get an indication how important this feature is for the participant.

How important is it or would it be if (feature) exists?

Evaluation

First,

• remove incomplete pairs, where not both questions have been answered and
• remove speeders that have an unrealistically short completion time

For evaluation purposes there are two more categories:

• Questionable: someone has not fully understood the described feature or question. This is the case if we get conflicting responses such as Like for both questions.
• Reverse: the proposed feature is the opposite of what the user wants. This is the case of someone says, he dislikes the functional version and like the dysfunctional version. In that case the participant is not interested in the offering at all and perhaps wants even the opposite.

Method 1: discrete analysis with the evaluation table

Step 1: Classification

Classify each answer pair to the six categories with the evaluation table.

• Functional: Feature present
• Dysfunctional: Feature absent

Step 2: Count

Simply count the number of participants per category.

Method 2: continuous analysis

The original Kano analysis uses discrete values. With discrete values, data in between is lost.

Therefore, William DuMouchel introduced a continuous calculation. He assigns a score to each functional and dysfunctional answer and considers reverse results and questionable data. Reverse response indicate that the users want exactly the opposite of what has been proposed. Questionable data refers to invalid or contradictory answers.

Step 1: Map each response to the correct values

Step 2: calculate the averages

Step 3: Map the feature to the graph

Closing remarks

Keep in mind that the Kano Model is a qualitative research tool. Therefore, don’t get misleaded by the numbers and use the Kano results alongside other research data.

Although the Kano Model will not be able to give you answers, why participants rate a certain feature with a certain category, it provides you a snapshot in time to understand what features bring what kind of value (satisfaction) to your users.