Editor's Introduction: When iterative product upgrades and user experience optimization, user research designers often need to borrow certain models for user analysis, so as to obtain clearer and more accurate user data. Among them, the KANO model and the PSM model are common analytical models. In this article, the author summarizes the methods and strategies of using the KANO model and the PSM model for demand analysis and pricing.
Both the KANO model and the PSM model are analytical models that user research and designers often use in their daily work to help define the prioritization and payment interval of user needs.
<h1 toutiao-origin="h2" >what is KANO? </h1>
The KANO model is a model for classifying and prioritizing user needs, based on the analysis of the impact of user needs on user satisfaction, and reflects the non-linear relationship between product performance and user satisfaction. It was invented by Tokyo Institute of Technology professor Noriaki Kano in the 1970s.
<h1 toutiao-origin="h2" >2, what is PSM? </h1>
The PSM model, also known as the Price Sensitivity Measurement model, is, as the name suggests, a model for testing the prices of products and services. The PSM price sensitivity analysis method was created by Van Westendrop in the 1970s.
<h1 toutiao-origin="h2" >3, how to use KANO and PSM models? </h1>
<h2 toutiao-origin="h3" >1. Usage scenario</h2>
KANO: Requirements analysis, such as before product development, we need to know which of these functions are basic functions, which are value-added functions, and how to prioritize functions, etc., can be defined by the KANO model.
PSM: When we determine the function of a good product, in the subsequent pricing, we can use the PSM model for price testing to seek the best price, the best price range and other information.
There are preset price ranges, optimized products or competing products, there will be approximate price ranges and references.
There is no preset price range, such as a complete new product, which requires a clear explanation of the product concept and an understanding of the cost of the solution for the consumer today.
<h2 toutiao-origin="h3" >2. Usage flow</h2>
Both are used in a way that results are presented from a clear purpose – questionnaire design – data processing.
Among them, there are two ways to set the price range in the questionnaire design in the price sensitivity test model:
Make a price gradient table, displayed as an option; however, it is limited by the price gradient between each option, which is not appropriate to use if it is in the high price of the product, such as the price of the car.
Some products can be set directly in the form of open questions due to more mature or competitive products, and the respondents have more freedom; of course, the minimum or maximum price limit can be set in the program to prevent extreme data.
<h1 toutiao-origin="h2" > fourth, KANO case use</h1>
<h2 toutiao-origin="h3" >1</h2>
The questionnaire settings divide the dimensions into two main aspects: satisfaction when provided, and satisfaction when not provided.
Each function/aspect needs to provide both positive and negative questions, and appropriate red/bold means can be used to prevent users from misreading.
<h2 toutiao-origin="h3" >2. Sample settings</h2>
In order to achieve a statistical sample size, it is recommended to quantify the sample size of more than 30.
In the subsequent cleaning of questionnaire data, some non-conforming samples may be cleaned, and the sample size can be increased appropriately.
In practice, this model may also be used in qualitative studies, i.e. sample sizes of more than 30 are no longer required.
<h2 toutiao-origin="h3" >3</h2>
Crossing this functionality and not providing it results in 6 different types of requirements:
A: Excitement/Glamour type;
O: Expectation/willingness type;
M: Basic/Essential;
I: Non-differentiator type;
R: Reverse/Reverse Type;
Q: Suspicious results.
According to the user's evaluation of a function, the proportion of this function in the above 6 classification attributes is obtained, for example, we get the XX function in A product to get its attribute O: expectation/willingness type, that is, when this demand is provided, user satisfaction will be improved; when this demand is not provided, user satisfaction will be reduced.
< h2 toutiao-origin="h3" >4</h2>
After deriving the attribute correspondence of each function, the better-worse coefficient can be calculated for the display of the four-quadrant chart.
Increased satisfaction coefficient Better= (A+O)/(A+O+M+I)
Eliminated coefficient of dissatisfaction Worse = -1 * (O+M)/(A+O+M+I)
For example, we get the coefficients for xx function in product A as follows:
Better=(26.7%+60%)/(26.7%+60%+0%+23.3%)=86.7%
Worse=-1*(60%+0%)/(26.7%+60%+0%+23.3%)=-60%
<h2 toutiao-origin="h3" >5</h2>
Calculate the better-worse coefficients for each attribute and then display them in the form of a four-quadrant chart.
Bounded by the better coefficient and the waves coefficient, respectively, and the midpoint, it is divided into:
The first quadrant is the excitement/charm type: if this demand is not provided, user satisfaction will not decrease; if this demand is provided, user satisfaction will be greatly improved;
The second quadrant is the expectation/willingness type: when this demand is provided, user satisfaction will increase; when this demand is not provided, user satisfaction will decrease;
The third quadrant is non-differentiator: whether it is provided or not, it has no impact on the user experience;
The fourth quadrant is basic/essential: when this requirement is not provided, user satisfaction is greatly reduced, but optimizing this requirement will not significantly improve user satisfaction.
In the actual daily work, we first have to meet the most basic needs of the user, that is, the basic/essential factors represented in the fourth quadrant.
After meeting the most basic needs, try to meet the user's expectation/willingness type of demand, that is, the functional requirements of the second quadrant, which is a competitive factor. Provide additional services or product features that users like to make their products and services superior to competitors and different, and guide users to strengthen a good impression of this product.
Finally, strive to achieve the user's excitement/charm type needs, that is, the first quadrant indication factor, to enhance the user's loyalty.
<h1 toutiao-origin="h2" >5, PSM case use</h1>
Mainly 4 questions, ask the respondents: too cheap to doubt the quality of the price not to buy, cheaper price, more expensive price, too expensive to buy the price.
The questionnaire can be selected as both optional and open question types, selected according to the range of prices and the desired range of precision.
<h2 toutiao-origin="h3" >2. Sample size</h2>
In order to achieve a statistical sample size, it is recommended that the sample size reach more than 30.
To solve the problem of respondents inflating or driving down prices, it is possible to increase the sample size and offset random errors.
If the questionnaire is conducted in the form of price options, you can also design price groupings, design 2 groups or several sets of price ranges through different price schemes, and conduct questionnaires to verify each other.
<h2 toutiao-origin="h3" >3</h2>
Downward cumulative statistics are performed on the price percentages of "too cheap" and "cheap", and upward cumulative statistics are performed on the percentages of "expensive" and "too expensive", and the four price lines shown in the following figure are obtained.
Among them, the intersection of "too cheap" and "too expensive" determines the optimal price, because in this case, the number of people who neither feel "too expensive" nor "too cheap" is the largest, and for the enterprise, at this price, there are the most consumers who are likely to buy his products.
At the same time, acceptable price ranges are determined by "too cheap" and "expensive", "cheap" and "too expensive".
Draw a line chart based on a statistical table:
<h2 toutiao-origin="h3" >4</h2>
The results can be presented using the line chart described above, expressing a reasonable price range of (3500, 5500) or even a smaller range (4000, 5000), with the ideal price point for maximizing the market at around 4500.
It is also possible to perform more accurate calculations, based on the intersection point of the calculated polyline, to obtain the following results:
Author: Mo Yu; Public account: Cool Jiale user experience design
This article is originally published by @Cool Jiale User Experience Design everyone is a product manager, and reproduction without permission is prohibited
The title image is from Unsplash, based on the CC0 protocol