Customer Insights Flashcards
Quality Dimension according to Garvin (1984)
Feature Reliability Conformance Durability PQ Serviceability Performance
Def Customer Insights
Customer Insights describe the market demands and customer expectations toward the product and the company.
What is used to enhance decision quality in data-driven product development
Digital Shadows
Dimensions of customer information
User
Environment
Usage
Types of customer information
Latent need vs Expressed need
Problem-oriented vs Solution-oriented
Direct customer feedback
Specific question
Collection of not yet existing data
Less data points
Indirect customer feedback
Targeted analysis of already existing data
Previously defined purpose
Big Data
Aim of Perceived Quality
Perceived Quality aims to transform human perception into a scalable quantity for the improvement of product development
Name the four steps of Stimulus Processing Chain
- Stimulus recording
- Multisensory stimulus processing
- Perception
- Action
Main Take-Aways from Today’s Lecture
The term Customer Insights covers the collection, analysis and
interpretation of customer data from heterogeneous sources. Customer
Insights are precise knowledge about the customer and his (latent) needs.
Perceived Quality aims to combine the conventional understanding of
quality and the customers perception.
The customer’s perception is a multidimensional process that companies try
to model and predict.
There are basic psychophysical relations between the intensity of stimuli
and their perception. How these features are combined to a quality
assessment, is a more complex research question.
Qualitative and Quantitative Description of Customer Data
- Subjectivity
- Degree of Structure
- Specificity
- Quantity
- Update Frequency
- Costs
Def Kansei Engineering
Kansei Engineering is an interdisciplinary product design approach. With Kansei Engineering, product stimuli are presented and customer Kansei descriptions are captured using questionnaires.
Steps of Kansei Engineering
- Selection
- Semantic Space
- Property Space
- Synthesis
- Validity test
- Modelling
Why Kansei Engineering?
Kansei Engineering is based on subjective assessments of products and helps customers express their (implicit) requirements for future products.
Kansei Engineering: The basic idea is to describe the product from two different perspectives
Semantic Space
Property Space
Principal Component Analysis (PCA)
- Assumption: variance of the variables is fully explained by the factors
- Goal: reproduce data structure as completely as possible with as few factors as possible
- The factors are called Principal Components
Principal Axes Factor Analysis (PFA)
- Assumption: variance of the variable is split between single residual variance and communalities
- Goal: discover communalities within the variables
- Estimation of the amount of communalities in advance
Def Reliability:
One measurement leads to the same results when performed several times at different times.
Def Validity
A measurement completely captures the underlying construct through the measured variable.
Cronbach’s Alpha
- indicates how high the internal consistency reliability of a group of items used to measure a characteristic is
- Cronbach’s Alpha is defined for -unendlich to +1. Only positive values can be interpreted meaningfully.
- If Cronbach’s alpha is close to +1, it can be assumed that the items are consistent and measure the same.
Name the modelling types of the kansei engineering
- Mathematical Models (Description of relationships in formulas)
- Iconic Models (Description of contexts in images)
- Verbal Models (Description of contexts in words)
Mathematical Modelling
Modelling circle
real problem –> Modelling –> mathematical problem –> Analysis Simulation –> mathematical solution –> Interpretation –> real Solution –> Checking –> real Problem
Main Take-Aways from Today’s Lecture - Exercise
Kansei Engineering is a methodology to systematically improve the
Perceived Quality of a product.
The reliability of customer clinic data can be evaluated with Cronbach’s
Alpha.
The decision tree can be used as a grey box approach to model the human
quality perception.
