How Subjective Clustering aids Affinity Diagram in grouping Customer needs in consumer products

(1)

How Subjective Clustering

aids Affinity Diagram in grouping Customer

needs in consumer products

SANDHEEP KUMAR VURUKKARA BOOPAL

(2)

How Subjective Clustering aids Affinity Diagram in

grouping Customer needs in consumer products

industry

Sandheep Kumar Vurukkara Boopal

Master of Science Thesis MMK 2016 MF228x KTH Industrial Engineering and Management

(3)

(4)

Examensarbete MMK 2016 MF 228x

How Subjective Clustering aids Affinity Diagram in grouping customer needs in consumer products

Sandheep Kumar Vurukkara Boopal

Godkänt Examinator Anders Berglund Handledare Susanne Nilsson Uppdragsgivare Creaffective GmbH Kontaktperson Florian Rustler

Sammanfattning

Insamling och analys av kundernas behov är viktiga delar i produktutvecklings- och innovationsprocesser. Dessa kundbehov måste vara i en form som lätt kan kommuniceras och förstås särskilt avproduktutvecklare i ett företag. Affinity Diagram är ett vanligt använt verktyg för att strukturera kundbehov. På grund av att metoden bygger på gruppdiskussioner, finns det risk för att enskilda individers åsikter inte tas tillvara. En metod som tar hänsyn till de individuella bedömningarna är Subjective clustering, vilkenhar utvecklats för att stödja Affinity Diagram.

Tidigare forskare har tillämpat båda dessa metoder i ett vetenskapligt och industriellt sammanhang och har funnit att det finns 92,5% av koppling mellan Affinity Diagram och Subjective clustering och drog slutsatsen att Subjective clustering stödjer Affinity Diagram. Det saknas forskning om huruvida Subjective clustering stödjer Affinity Diagram för konsumentprodukter. För att undersöka detta har en fallstudie på konsumentprodukter i ett produktutvecklingsprojekt i Creaffective GmbH genomförts.

(5)

(6)

Master of Science Thesis MMK 2016 MF228x

How Subjective Clustering aids Affinity Diagram in grouping customer needs in consumer products

Sandheep Kumar Vurukkara Boopal

Approved Examiner Anders Berglund Supervisor Susanne Nilsson Commissioner Creaffective GmbH Contact person Florian Rustler

Abstract

Collection and analysis of customer needs are important parts of product development and innovation processes. These customer needs must be in a form that can be easily communicated and easily understood especially by the R&D personnel. Affinity Diagram is one such tool to structure these data. Because of the nature of the Affinity diagram method, it is prone to biases. An alternative method that exists is Subjective clustering. It has been developed as an aid to support affinity diagram.

Previous researcher has applied both these methods in a scientific and industrial context and has found that there is 92.5% of association between affinity diagram and subjective clustering and concluded that Subjective clustering aids affinity diagram. However there has been no research on whether subjective clustering aids affinity diagram in consumer products context. Taking this as a research gap, this thesis is performed, taking the Innovation project at Creaffective GmbH, as a case study.

(7)

(8)

FOREWORD

This Master thesis concludes my Master study of Integrated Product Design (Product Innovation Management) at KTH Royal Institute of Technology. Over this process of this Master thesis, there has been a number of people who have been a very important part of this Master thesis, and the reason as to why the study has been completed and stands in its place.

First and foremost, I would like to heartily thank Mr. Florian Rustler, founder of Creaffective GmbH, Munich for proposing the Innovation project that served as the right platform to conduct this study and also shepherding the project, guiding me at the right and crucial stages and providing the right contacts. I would also acknowledge the participation of Isabela Plambeck and Daniel Barth in the study and also assisting in the various stages of the project.

The study would not have been possible without the participation of Mr. Eashwara Krishnan, Mrs. Subashini Eashwar, Mrs. Supriya Satish, Mr Sarma, Mr. Velpari and Mrs. Bhuvaneswari who are trained trainers at Junior Chamber International (JCI), and also JCI without which I would not have known them. Their participation is definitely invaluable.

I must definitely acknowledge the participation of several people who have been the source for various inputs for the study. To start with, Mr. Subramanian, Past National President of JCI India; Mr. Mohammed Nassar, Founder E2E Excite; Baalachandran Gopinath, HRD Trainer, India; Dhananjaya Hettiararchi, HR Trainer, Sri Lanka and word champion in public speaking 2014; Ankur Grover, Founder, Tinker Lab, India; Mr. Harsha, Freelance Trainer, India; Mr. Sivakumar Palaniappan, Career Coach, Masteringmind Academy, India; Mr. Jayaraman Umashankar, founder, Karna communication academy, India; Mr. Jim Clark, Design Thinking trainer, Innogreat, Taiwan; Mr. Chendil Kumar, CK consultants, India; Mr. Randy J Harvey, Keynote speaker, Bassinger & Harvey, US; Dali Han, Bosch China; Ms. Sandhya Sridhar, Mercedes Benz, India; Mr. Naveen Ramkumar, Robert Bosch, India; Mr. Satish Ramachandran, SME Head, Vodafone, India; Mr. Arjun Murali, General Electricals, India; Mr. Krishna Devarajulu, US Bank, US; Mr. Nithin Joseph, Manager, Taj Vivanta, India; Mrs. Grace Meng, Jingling Hotel Nanjing, China.

Finally, I would like to express my sincere gratitude to both my supervisors Mats Magnusson and Susanne Nilsson who especially guided me during the thick and thin. My gratitude to my alma mater, KTH Royal Institute of Technology for providing me the required knowledge and making me who I am to make this happen.

(9)

(10)

NOMENCLATURE

Abbreviations

AD Affinity Diagram B2B Business to Business B2C Business to Consumer HC Hierarchical Clustering

JCI Junior Chamber International, an International organisation

JTBD Jobs To Be Done

KJ Method Also called as Affinity Diagram, founded by Jiro Kawakita

OEM Original Equipment Manufacturer

PDT Product Development Team

PET Poly Ethylene Terephthalate

QFD Quality Function Deployment

SC Subjective Clustering

TELCO Telecommunication

TMR Traditional Marketing Research

(11)

(12)

1 INTRODUCTION

This chapter describes the background, the purpose, the limitations and the method(s) used in the presented project.

1.1 Background

Identifying the right customer needs is a very crucial task for any company involved in New Product Development. Research shows that companies spending more resources and giving more importance to this activity in the initial phase of innovation, performs better in terms of many attributes such as customer satisfaction, market share and profitability (Cooper & Kleinschmidt, 1993; Rothwell, 1992). Further, Beyer & Holtzblatt (1999) showed that the best product design results when the product designers are involved in collecting and interpreting customer data and appreciate what real needs of the customers are. However, this is not an easy task. A number of methods have been designed and proposed for this task in the literature, by various academics and researchers. Structuring the data collected is an important phase of the need analysis phase, as this is the phase that gives the direction for the product developers to create the product (Griffin & Hauser, 1993). Affinity Diagram (AD) is a popular tool in organizing the chunks of data to move from an abstract level to concrete understanding of the data, as they organize ideas into categories based on their underlying similarity (Shafer, Smith, & Linder, 2005).

1.1.1 Previous Work

Many researchers agree that AD is prone to bias. Since it involves the discussion of many people (usually around 4-5) and each adds their own point of view, usually there are situations where-in some of the members dominate others and it is usually agreed, although there would not be any real agreement. For example, Takai & Ishii (2010), (p.102). say that “few participants’ opinion may skew the results”. To overcome this drawback, , it was demonstrated that the use of Subjective Clustering (SC) may aid to support affinity diagram(Takai & Ishii, 2010). Subjective Clustering (SC) is an alternative grouping method which is based on statistical analysis of individual’s grouping.

(15)

1.2 Purpose

The existing literature of comparison of Affinity Diagrams and Subjective clustering in grouping customer needs is very limited. In addition, the research performed was conducted in a highly scientific context. The result may hence not be applicable in consumer products context as, the industry and target customer genre is completely different. For e.g., it may be dubious to use the results of the previous research in grouping customer needs in consumer products such as mobile phones or daily durables. Hence, there is a research gap, which is to verify if SC supports AD in grouping customer needs in consumer products industry. Or in other words, to verify if there is a good match between the grouping results of AD and SC. Consumer products industry is specifically an interesting context to perform this study in , as this is an industry that is characterized by continuously changing demographics and consumer preferences (Renner, 2016). Hence there arises a need to find if SC aids AD in consumer products industry segment.

In addition, in the previous research, the grouping of the needs was done by the product development team. Any bias in this process, multiplies in the forthcoming stages of the product development, finally leading to products that become inappropriate for the customers (Takai & Ishii, 2010). Many researchers agree that participation of customers in the product development process adds more value and provides positive benefits (Morgan & Obal, 2016). This provides a motivation to verify the usage of SC as an aid for AD, by comparing both, Product development team’s and customer’s results.

The study presented in this thesis is being carried out in a project at Creaffective GmbH, an innovation consulting company, based in Munich, Germany. The project is an explorative attempt to find if there are opportunities to create better training aids and equipment (such as Flipcharts, White boards etc.) for innovation workshop facilitators, trainers in the South Asian markets such as India and China. It hence, provides an opportunity to understand how the tools to handle the structuring of customer need is influenced when used in a consumer product context.

Hence this thesis is aiming to answer the following research questions,

1. To what extent is Subjective Clustering aiding Affinity Diagram in grouping customer needs in the context of consumer products?

2. What could be learnt by comparing the results of Affinity Diagrams of Product Development Team and customers; Subjective clustering of Product Development Team and customers?

(16)

Fig 1 Research Design

1.3 Method

To answer the research questions, it seems logical to use the same procedure as followed by the previous researcher, so that the results are more consistent and there wouldn’t be any bias because of the methods that would be used. A mix of both qualitative and quantitative methods was used.

The first step was to collect the condensed customer needs from the relevant customer segment. This formed the qualitative part of the study research and it involved interviewing customers about their needs. This seemed to be the best method as it allowed for exploration on what the underlying needs of the customers were.

Secondly, the needs were first grouped by the members of the product development team and the target customers individually, i.e. a Subjective clustering was performed. The third step was to group the needs as a team i.e using Affinity Diagram. So the product development team grouped the needs together and as did the target customers. The fourth step was to compare the results of AD and SC, by the product development team and compare this results of AD and SC, to that of the target customers. This was achieved by utilizing a parameter called Goodman-Kruskal’s lambda, which is the same parameter utilized in previous research studies (Takai & Ishii, 2010)

The fifth step was to compare the AD results of the product development team and the group of target customers. Due to the nature of the way AD is carried out, the author participated in the process by taking a neutral stand and observed the Affinity Diagram process.

(17)

1.4 Delimitations

This study is based on a single Innovation project, limiting the number of data analysed. Also, the product development team consisted of three Innovation coaches and one student (the author). The members of the team have hence large differences in their prior experience in real product development which may have influenced the result of the understanding of the customer needs.

The customer needs are collected through a traditional marketing research technique (TMR), which is speaking to customers. One drawback about this technique is the unreliable memory of the customers. The customers do not completely remember their past experiences during the interview (Price, Wrigley, & Straker, 2015).

All the interviews were done through telephone conversation because of the geographical limitation of the thesis project. Hence there were no opportunities to observe other important drivers of the quality of the research such as participant behavior, body language and their facial expressions.

(18)

2 FRAME OF REFERENCE

The reference frame is a summary of the existing knowledge and former performed research on the subject. This chapter presents the theoretical reference frame that is necessary for the performed research, design or product development.

For a clear understanding, some of the basics are explained in this chapter, such as what customer needs are and what is unique with the consumer segment, followed by literature related to the importance of structuring customer needs. Literature concerning the specific tools in use for the study, Affinity diagram and Subjective clustering is presented. Since the Subjective Clustering deals with clustering aspects, some light is thrown on these aspects, with special reference to Hierarchical Clustering.

2.1 What are Customer needs and why should it be

structured?

There has always been a debate on what the meaning of customer needs are and what exactly defines a customer need statement. In simple words, customer buys various products and services to “get their job done”. This leads to the concept of “Jobs to be done” (JTBD), which can be associated with customer needs. JTBD are not products or services or a specific solution but a higher purpose as to why a customer shall buy a particular product or service (Silverstein, Samuel, & DeCarlo, 2013). When looking at the markets through the “jobs to be done” lens, a customer need statement is best defined as what the customer measures as the success and value when getting a job done (Ulwick, 2016). The better the customer needs are understood and documented, the better the product developers will be able to make informed decisions in their design work (Patnaik, 2009).

(19)

Fig 2. Jobs to be done Breakdown (Silverstein et al., 2013)

Experts agree that customer’s need is the core of the product and value offering to customers. Satisfying them is the primary task of the executives (Fan & Cheng, 2006). Several studies have been conducted in proving the importance of understanding the customer needs in the early phase of new product development and innovation. Voice of the customer (VOC) is a term used in business and Information Technology to describe the in-depth process of capturing customer’s expectations, preferences and aversions. One famous study done by Cooper(2001), suggests that capturing the VOC, doubles the success rate of the new products and have more than 70% higher market share compared to the ones who have a poor approach.

Importance of customer needs can also be seen in the area of Quality Function Deployment (QFD). Griffin & Hauser (1993) mention that there are three steps in QFD’s customer input: 1. Identifying customer needs 2. Structuring customer needs 3. Setting priorities for customer needs. The overall goal of QFD is to help the product development team understand how to satisfy the customer. It was noted that the usage of QFD results in a 60% reduction in design costs and 40% reduction in design time (Hauser & Clausing, 1988).

The collected customer needs and information is usually unstructured and contains more than what is necessary. So it becomes important to reduce the information to a manageable amount of data. The data from one customer gives depth regarding the subject, however only analyzing a group of customer’s needs, shows a pattern of the bigger picture (Vanalli & Cziulik, 2003). Need assessment also refers to structuring and analysing customer needs for the reason of developing new products and services. The information on customer needs, to be useful, must be in a form that can be easily communicated especially to the persons in R&D who need it. (Adams, Day, & Dougherty, 1998; Gupta & Wilemon, 1988). The grouping and structuring of collected needs can be done by three methods, viz., Affinity Diagram, Relationship diagram and tree diagram (Shillito, 2000; Burchill & Brodie, 1997; Mazur, 1997).

2.2 Consumer products – What is different when it

comes to analysing customer needs

(20)

consumers) segment and B2B segment (for example glucose bottles for drips in hospitals). If it is the case of B2C, they might have to test their prototypes with many customers with respect to demography, age or aesthetic aspects. For e.g., water bottles designed for children would be different from that of the adults and would again be different from that of old people. The company might have to create prototypes after each feedback loop to test them again. However in case of B2B, the company might just have to follow certain standards and norms (may be in terms of material, sizes etc), a fixed blueprint and meet them. In this case, the customer shall be a glucose bottle OEM.

According to Day, Schoemaker, & Gunther (2004), the emerging technologies and competition is forcing to reduce the product development time than it was earlier in consumer product industry. In consumer product industry, the developers cannot afford to lose contact with the customers, and the customers always play the central part of the product development process.

Fig 3. Two paradigms in industrial products and consumer products (Von Hippel, 1979)

Von Hippel, (1979), who is considered to be the father of user centered innovation, places a lot of importance in the user being the main catalyst in the innovation and product development in the consumer product segment. These days, the majority of the companies have realized the importance of having a customer in the product development team, and developing new products along with them. Fig 3 shows the two paradigm differences between consumer and industrial products. In consumer products, the customer has a major say on the idea screening, selection and analysis. Provided these differences and contrasting features between new product development processes in consumer products and industrial products, it is a clear argument that the research work done by (Takai & Ishii, 2010) cannot be directly extended to concern also consumer products.

2.3 Affinity Diagram method

In section 2.1, customer needs were discussed along with their importance and why they must be structured.

(21)

various steps involved in AD as adapted from ( Awasthi & Chauhan, 2012; Burge, 2011) is as follows:

1. Make the problem clear and transparent to all the team members 2. Each team member should write one idea per post-it.

3. These post-its must be placed in a way, which should be visible to all the team members.

4. Each team member comes to a consensus for grouping each idea by brainstorming if the chosen need (idea) matches with any of the other ideas. Once everyone agrees, the grouping process is stopped.

5. A header title is provided to each group and a final structured affinity diagram is created.

AD is a very useful tool in many applications. It is a consensus based approach. As helpful as it may sound, there are a few down sides to the method. Situations sometimes arise when there will be few group members who are dominant and hence subsiding the contributions of the rest of the group members. The way AD is designed does not accommodate a solution to overcome this bias. So there is clearly a need of a method where everyone contributes equally (Takai & Ishii, 2010).

2.4 Subjective clustering method

Subjective clustering is one solution to this problem. It is a statistical tool that ensures that all team member’s knowledge is equally considered. ((Griffin & Hauser, 1993; Takai & Ishii, 2010). It is based on the statistical analysis of the grouped results of the individuals of the team. With reference to Fig 4, the various steps involved in Subjective clustering as adapted from (Green et al., 1969) is listed below:

1. Each individual group the needs based on similarity

2. A similarity matrix is constructed based on the individual’s groupings. In a similarity matrix, 1 is assigned to an mth row and nth column when ideas m and n are grouped together. Otherwise 0 is assigned to all the positions. The diagonal elements of the similarity matrix are always 1 because each customer need is grouped with itself.

3. A co-occurrence matrix is constructed by adding all the similarity matrices. An mth row and nth column in a co-occurrence matrix tells how many individuals have grouped the m and n ideas together. So the larger the number, it means that more are the people who have grouped those elements together. The entire diagonal element in a co-occurrence matrix is the same and it tells the total number of team members. 4. A Dendrogram is constructed using a Hierarchical Clustering procedure (HC) Johnson (1967). In Dendrogram, the more similar needs are grouped at the lower level and the less similar ones are grouped at the higher level.

(22)

Fig 4. Subjective Clustering (Takai & Ishii, 2010)

2.6 Clustering

Clustering is the backbone of SC. Hence some background to understand what it means, what kinds of clustering methods are available and which method is used in this study is discussed in this section.

Cluster analysis or clustering as it is called is the task of statistically grouping a set of objects in such a way that the members of one group are more similar to each other than the members of the other groups. There are a number of clustering algorithms available such as K-means, Hierarchical clustering, Fuzzy C-means and mixture of Gaussians. Hierarchical clustering is used in this particular thesis . The reason is that in other methods such K-means or Fuzzy C-means, the number of clusters must be mentioned in the beginning, which basically takes away the freedom of analyzing the dendrogram (tree structure) and manually finding the optimal number of clusters. Hierarchical clustering is one method which gives these benefits and completes the meaning of using a dendrogram.

2.6.1 Hierarchical Clustering

(23)

In general, all the clustering algorithms fall under two types of clustering namely, agglomerative clustering and divisive clustering. In agglomerative clustering, each observation is assigned to its own cluster. Then the similarity (or distance) is computed between each of the clusters and the two most similar clusters are joined. These steps are repeated until one single cluster is left. However the divisive method is the opposite of agglomerative clustering where all the data points are associated with one single cluster and then it is partitioned to two least similar clusters. These steps are repeated until there is one cluster for each observation. Fig 5 shows the pictorial representation of the two types of hierarchical clustering.

Fig 5. Agglomerative and Divisive hierarchical clustering (Sayad, 2016)

(24)

(25)

3 RESEARCH STUDY DESIGN

In this chapter the working process is described. A structured process is often called a method and its purpose is to help the researcher/developer/designer to reach the goals for the project.

This thesis followed a chronological process as displayed in Fig 7.

(26)

3.1 The innovation project studied

The study is being carried out in a project at Creaffective GmbH, an innovation consulting company, based in Munich, Germany. The project is an explorative attempt to find if there are opportunities to create better training aids and equipment (such as Flipcharts, White boards etc.) for innovation workshop facilitators, trainers in the South Asian markets such as India and China. Although this project is more extensive and goes to the extent of developing ideas and realizing them, this master thesis makes use of the data collected in the first phase in the project.

There are five core team members involved in the product development. Three being innovation coaches and facilitators and one Industrial Designer and one being a student of KTH Royal Institute of Technology (the author). Four of these, excluding the Industrial Designer were involved in the research project. These four are involved in the initial grouping of needs (AD) and later individually (SC). The author also performed the customer need collection.

3.2 Brainstorming and filtering questions to get

relevant answers

Griffin & Hauser (1993) recommends based on their findings that for identifying the needs in consumer products, one-on-one interviews are more cost effective than focus groups. They suggest that 20 to 30 interviews (customers) are necessary to get 90-95% of the customer needs and multiple analysts or team members should read and interpret the raw transcripts. These suggestions were followed in the case study of this thesis. The Product Development team at Creaffective GmbH brainstormed a total of 78 questions to ask the various customer profiles. The team later came to a consensus to ask the most relevant questions (see the sample interview guide and the response in the Appendix). Shortly describing, the questions contain categories such as problems the trainers face and what they do on such events. On an average each respondent were asked around 20 questions and the interview was held for around 30 minutes. Notes were taken while simultaneously interviewing the respondents.

3.3 Customer Interviews

(27)

what he meant, the problems he faced and how usually he fixes them with temporary solutions.

Fig 8. One interviewee demonstrating the issues he has during a training program

3.4 Cleaning the data and deducing the list of

customer needs

(28)

Fig 9. All the collected needs are laid on an open space to get a first-hand look

3.5 Subjective Clustering and Affinity Diagram

“Customers” in this thesis means the end customers who are likely to use the new products that are expected outcomes from the project at Creaffective GmbH. For the study, a different set of customers from whom the customer needs were collected, but with the same customer profile is being used, for the simple reason being, to avoid the bias of grouping their own needs. The customers for this part of the study are chosen from the organization, Junior Chamber International India, an organization dedicated to self-improvement and training. A total of six participants were considered under the category of customers. All the participants are national level graduates from the organization, with about five to ten years of experience in training. To get more buy-in from the participants, an amazon gift card of worth $10 was provided to each of the participant.

Customers were first asked to individually group the customer needs. A similarity matrix was constructed, followed by co-occurrence matrix (referring to section 2.6). By performing hierarchical clustering, and through generation of dendrogram, the needs are grouped (SC). Through a TELCO (Tele-Communication), all the six members grouped the needs together by following the AD principle and this process was performed in about an hour.

(29)

Fig 10. Final grouped needs

3.6 Comparison and Analysis

3.6.1 Comparison of SC and AD of respective groups

The first research question relates to the supporting role of SC in consumer products. The answer shall be found in the comparison of SC and AD by both customer and PDT groups separately. So Goodman and Kruskal’s lambda is being used for this purpose. The motivation to use this parameter is that, this is the same parameter that was used in the previous research, and hence there will be compatibility in comparing the previous results and the new results. Goodman-Kruskal lambda is used to measure association of the cross tabulation of nominal level variables.

λ = (S-R)/(N-R) where,

λ is Goodman-Kruskal’s lambda

S is the sum of the highest cell count for each row

R is the highest row total

N is the total of all cell counts (Minitab, 2016)

3.6.2 Comparison of AD results of customers and PDT; Comparison of SC results of customers and PDT

(30)

discussion styles, thinking style based on the categorical names given by the participants were analyzed.

Comparison of SC of both PDT and customers is however done through quantitative comparison. A special package called “cluster.stats” was used to analyse the individual clustering. The package provides results based on various parameters (refer Appendix E) and then they are analysed.

(31)

4 RESULTS

In the results chapter the results that are obtained with the process/methods described in the previous chapter are compiled, analysed and compared with the existing knowledge and/or theory presented in the frame of reference chapter.

4.1 List of customer needs

Appendix C displays the final summary of customer needs collected through the various interviews in the study.

4.2 Comparison of grouped needs by AD and SC by

Product Development Team

In appendix H the individual groupings of the customer needs by the product developers is presented. All the individual responses were converted into similarity matrix, using the rules described in section 2.4. Fig. 11, shows the construction of co-occurrence matrix. Adding all the similarity matrices gives rise to co-co-occurrence matrix.

Fig 11. Co-Occurrence Matrix of PDT

(32)

Fig 12. Dendrogram from SC by PDT

Refer Appendix D for the tabulated groups of customer needs from SC and AD. Table 1 is created by comparing and contrasting the needs of AD and SC. Four comparisons are obtained by best fit (Since there were ten groups formed under AD and only four under SC, the nearest groups are combined together to compare and contrast with the four groups of SC). The colors indicate the respective customer needs that do not fall under the same comparison group.

Table 1. Comparison of AD and SC by PDT

Ambience issues and Venue rules

Training logistics, format and participant learning

issues

Technical, equipment issues and wishes

Workarounds, trainer preferences and client’s

(33)

28 29 29

34 34

46 46

For the same reason, to measure the degree of similarity or association, Goodman and Kruskal’s λ is calculated by using the formula (refer section 3.6.1). Refer Table 2, The elements common among both AD and SC are filled along the diagonal and the un-matched elements are associated with that particular category. For e.g., in the first group, there are 14 elements that are common to both AD and SC, hence 14 is filled in the first row and first column. The need number 28 (green) is an un-matched element in SC in the first comparison group and an un-matched element in AD in third group. So in the table, since there is only one un-matched element, it is given to first group under SC and third group under AD. In the similar way, the entire table is constructed and the index is calculated to be 0.709

Table 2. Association between AD and SC grouping results of PDT

SC A B C D AD A 14 14 B 8 1 9 C 1 12 4 17 D 2 1 5 8 15 10 13 10

4.3 Comparison of grouped needs by AD and SC by

Customers

See Appendix I for the individual grouping results of the customer needs by all the six customers. All the individual responses were converted into similarity matrix, using the rules described in section 2.4. Fig 13, shows the construction of co-occurrence matrix. Adding all the similarity matrices gives rise to co-occurrence matrix.

(34)

Using R program, (refer Appendix F) a distance matrix is constructed and using hierarchical clustering, a Dendrogram is generated. Arbitrarily, the dendrogram is cut at a certain height (in this case where it gives four clusters). Refer fig 14.

Fig 14. Dendrogram from SC by Customers

Refer Appendix D for the tabulated groups of customer needs from SC and AD. Table 4 is created by comparing and contrasting the needs of AD and SC. Four comparisons are obtained by best fit (Since there were nine groups formed under AD and only four under SC, the nearest groups are combined together to compare and contrast with the four groups of SC). The colors indicate the respective customer needs. There are some needs that do not fall under the same comparison group.

Table 4. Comparison of AD and SC by Customers

Clients attitude, participant’s attitude and trainer’s technique

Training planning, aids, setup and Trainer’s attitude

(35)

40 40 16 16 22 23 12 42 42 20 20 27 27 24 24 44 44 25 25 28 29 29 39 26 26 35 35 34 34 30 30 43 43 46 46 31 31 47 47 32 9 33 33 36 1 37 37 39 13 41 41 45 45 48 48 7 17 23 22

For the same reason to measure the degree of similarity or association, Goodman and Kruskal’s λ is obtained by using the formula (refer section 3.6.1). The elements common among both AD and SC are filled along the diagonal and the un-matched elements are associated with that particular category. For e.g., there are 7 elements that are common to AD and SC in the first group and hence 7 is written at the first row and first column. The need number 36 (blue) is an un-matched element in SC in the third comparison group and an un-matched element in AD in second group. So in the table, since there is only one un-matched element, it is given to third group under SC and second group under AD. In the similar way, the entire table is constructed and the index is 0.58

Table 5. Association between AD and SC grouping results of customers

SC A B C D AD A 7 1 8 B 2 15 1 1 19 C 5 6 11 D 2 8 10 9 23 7 9

4.4 Comparison of grouped needs of both AD and SC

by Product Development Team and Customers

(36)

Table 6. Comparison of AD by PDT and customers AD PDT Customers 9 32 28 27 24 2 24 20 43 21 30 35 15 40 37 21 1 39 31 47 5 39 12 7 35 10 6 11 34 32 46 16 47 40 16 43 8 38 45 8 36 42 9 18 29 42 34 14 27 2 1 36 46 3 17 10 22 38 13 12 4 29 33 18 44 14 19 26 19 41 11 20 44 6 48 13 7 23 15 48 5 31 4 25 28 26 30 17 23 33 3 37 41 45 22 25

Table 7. Comparison of SC by PDT and Customers

Parameters PDT Customers

Number of cases 48 48

Number of clusters 4 4

Vector of cluster sizes (number

of points) [15 10 13 10] [23 9 9 7]

Size of smallest cluster 10 7

Number of noise points 0 0

Vector of cluster diameters [5.86 8.10 8.72 9.47] [13.97 9.58 9.02 7.36] Within cluster average distances [2.91 5.18 4.67 6.13] [9.44 6.80 6.72 5.13] Within cluster distance medians [3.53 5.50 4.56 6.29] [9.58 7.33 6.54 5.30] Separation [7.29 6.37 3.95 3.95] [6.21 6.21 6.85 6.62] Average toother 14.47 11.91 12.43 10.57 [12.43 13.16 12.91 12.44] Separation Matrix 0.00 8.36 8.48 7.29 8.36 0.00 7.97 6.37 8.48 7.97 0.00 3.95 7.29 6.37 3.95 0.00 0.00 6.21 6.85 6.62 6.21 0.00 7.84 10.66 6.85 7.84 0.00 8.48 6.62 10.66 8.48 0.00 Matrix of mean dissimilarities

(37)

15.98 11.95 0.00 7.58 13.69 9.79 7.58 0.00

11.34 14.89 12.82 0.00

Average distance between

clusters 12.49 12.70

Average distance within clusters 4.31 8.62

Number of distances between

clusters 855 782

Number of distances within

clusters 273 346

Maximum cluster diameter 9.47 13.96

Minimum cluster separation 3.95 6.21

Within cluster sum of squares 585.290 1509.59 Cluster average silhouette

widths 0.76 0.45 0.37 0.16 0.07 0.43 0.44 0.53

Average silhouette width 0.47 0.28

Goodman and Kruskal’s

Gamma coefficient 0.95 0.75

G3 coefficient NULL NULL

Pearsongamma 0.74 0.60

Dunn index 0.41 0.44

Minimum average dissimilarity

between two clusters 1.23 1.20

Entropy of the distribution of

cluster memberships 1.37 1.26

Ratio of average distance within clusters and average distance

between clusters

0.34 0.67

Calinski and Harabasz index 63.51 17.51

Vector of widest within-cluster

gaps 4.79 4.45 4.56 5.10 7.50 6.77 6.29 5.40

Widest within-cluster gap 5.10 7.50

Separation index 4.05 6.39

Corrected Rand Index NULL NULL

Variation of Information Index NULL NULL

4.5 Importance of needs

(38)

(39)

5 ANALYSIS AND DISCUSSION

A discussion of the results that the author has drawn during the Master of Science thesis is presented in this chapter.

5.1 Discussion

The United States Consumer Product Safety Act (CPSA), enacted in 1972 by Congress, has an extensive definition of consumer product: "any article, or component part thereof, produced or distributed (i) for sale to a consumer for use in or around a permanent or temporary household or residence, a school, in recreation, or otherwise, or (ii) for the personal use, consumption or enjoyment of a consumer in or around a permanent or temporary household or residence, a school, in recreation, or otherwise; but such term does not include— (A) any article which is not customarily produced or distributed for sale to, or use or consumption by, or enjoyment of, a consumer". As per the above definition, the products that are studied in this thesis can be classified as consumer products. Further it can be noticed that the customer needs collected from the customers had resemblances with the format of JTBD. For e.g., consider customer need #24, “I want a better clicker, as the current ones are bulky and when I place them on the pocket, it bulges out”. The social dimension of the related jobs to be done can be seen. On the other hand, consider customer need #29 which says, “I wish to have a high tech transparent screen and voice control”. Here, the personal dimension of the main jobs to be done can be seen. Customer need #15 says, “I wish I was able to quickly control the brightness of the lighting in the training hall”, which describes the functional aspects of the related jobs to be done. Customer need #45 touches on the emotional aspect of the main jobs to be done, which says, “Sometimes the laptops and projectors are not compatible at all and it is very frustrating”. So in this way it can be seen that in reality, the needs vary differently from customer to customer and this could be once again taken as a strong motivation as to why we need a similar research in the consumer products industry.

(40)

5.1.1 Comparison of grouped needs by AD & SC by PDT

Directly comparing the AD and SC results of PDT through Goodman and Kruskal’s lambda it yields a value of 0.709. In other words, it can be understood that the grouping needs by AD method and is 70% similar to SC or vice versa.

Conversely it can be thought that there is 30% dissimilarity (about one-thirds, which is significant). One argument as to why 70% is a reasonable value is as follows. In B2B in general, it can be regarded, there is just one customer (in fact many but all the customers look for the same solution), but in B2C, there are innumerable number of customers. Which also means that when there is just one customer, the diversity in the needs of the customer is reduced. However, when it comes to B2C, diverse customer needs exist. So what could have likely happened is that every diversified need is interpreted in a different way by every product developer in SC as it is famously said a product developer is to think by putting themselves in the shoes of the customer (Whittle & Foster, 1989). This could be compared and contrasted with the previous research done by (Takai & Ishii, 2010), where 92.5% similarity or association was found which was in a B2B setting or an industrial/scientific context. This could be understood in the light of the reason that in scientific context, because of the nature of the project, all the scientists must be well informed, and be extremely clear about every detail of the project, which gives a high similarity index compared to consumer products segment.

5.1.2 Comparison of grouped needs by AD & SC by Customers

(41)

On further examination of the individual groupings, it was found that the category names under the AD were more specific than generic like the ones under SC. The names given under individual groupings resembled more like a primary need (for e.g., Attitude, flexibility, comfort, constraint and challenge). However the category names on AD resembles more like a secondary need (for e.g., Client’s attitude, participant’s attitude, Trainer’s attitude), perhaps thanks to some discussion that happens. Griffin & Hauser (1993) defines primary needs as those top needs that give the strategic direction for marketing. Each primary need further elaborates into secondary needs. Each secondary need indicates specifically what the marketing manager should do to satisfy the corresponding primary need. This, in a way is helpful that, later it helps the product developers during the product development phase to be more concise in certain features of the product.

It was observed that having more discussions lead to a more refined and more pragmatic clustering of the needs under appropriate headings. Otherwise, just considering results from SC wouldn’t have given any meaning at all.

As the moderator, the author of this paper felt that there was a methodology missing in moderating an AD discussion. There is a need to identify additional methods to better support the group discussions.

5.1.3 Comparison of grouped needs by AD by PDT and customers

There weren’t any significant implications on comparing the AD of PDT and customers except that it was noticed that there is a difference in thinking styles of the product development team compared to that of the customers. For instance, (refer to need #16), where it says, “Sometimes I have to stick two smaller papers to create a bigger paper to stick it on the wall to place participant's work”, it was interpreted as a workaround (Workaround means a method for overcoming a problem or limitation in the system) by the product development team and it was also unanimously agreed to group it alone. In contrast, the customers grouped it under the Training aid category. There was more deliberation between the PDT on whether to categorize it as “Workarounds”, although it was the only category with only one need. This indicates that the product developers are prone to become more solution oriented than the customers. That may not be as surprising as the goal for a product developer is to identify and design solutions for their customers.

5.1.4 Comparison of grouped needs by SC by PDT and customers

(42)

understanding that the Product Developers choose to group needs that are more similar to each other compared to the customers, which in turn hints that there is more similarity in thoughts of the product developers compared to that of customers. Or it can also be interpreted as that the product developers think about the solutions for the problems for customers as a whole and each of the customer thinks and groups the customer needs from their own perspective, causing the points inside the cluster to be further away.

But the average distance within clusters in PDT category is much lesser, in fact exactly half of the customer category. This leads to an understanding that the cluster members in customer segment are further away compared to PDT category. Or in other words, the members of a cluster in PDT are more similar compared to the members in a cluster in customer segment. Perhaps, the PDT has a more unified understanding and are on the same page whereas, the needs are different for every customer, even though they all belong to the same category, re-iterating the first implication made.

(43)

Fig 16. Symbolic representation of the spread of clusters and customer needs in SC

(44)

6 CONCLUSION

The conclusions are based from the analysis with the intention to answer the formulation of questions that is presented in Chapter 1.

In conclusion, this thesis has brought to light the importance of AD and SC in structuring the customer needs, after the work performed by Takai & Ishii ( 2010). It is surprising to see that not much work has been done in this very specific field of research. It is hoped that there shall be more experiments conducted in the days to come to help the academia and the industry to better understand this important task in the product development process.

The first and foremost aim of the thesis was to find to what extent the SC supports AD in consumer products. The answer to this question is given by the comparison between SC and AD of customers and between SC and AD by PDT. In the table 8, the research questions are presented along with the respective conclusion.

Table 8. Conclusions

S.No Research Question Conclusion

1

To what extent does Subjective Clustering aid Affinity Diagram in grouping customer needs in a consumer products context?

There was 70% association between AD and SC of PDT. This is much lesser than the 92.5% association obtained in the previous research. The PDT’s results are higher than the customer’s which stands at 58%. In both the instances it was found that the results were lesser compared to the existing research. This comes at no surprise because of the reason that in consumer products industry, the needs of the consumers are so diversified, that each consumer has different preferences. It can also be seen from the lesser match value among the customers compared to the product developers. Further it was found from the comparison of AD and SC by customers, that having more discussions lead to pragmatic clustering of needs. So more of AD is recommended in consumer products context as they give more specific results. Hence this is a motivation for the Product Development Team to organize more intensive focus group discussions to generate further insights and structure the needs in a better way. Subjective Clustering seems to be more suitable in industrial context, such as the example in previous research, where the customer needs are consistent across all players and not diversified.

2

What could be learnt by comparing results between subjective clustering of Product Development Team and customers?

(45)

7 RECOMMENDATIONS AND FUTURE WORK

In this chapter, recommendations on more detailed solutions and/or future work in this field are presented.

The one biggest take away from the thesis work would be that, in the consumer products industry, usage of Affinity Diagram methods would lead to better results. So it is recommended for managers in companies to concentrate more on Affinity Diagram methods for structuring the customer needs for consumer products. However, during the course of the research, it was observed that more disciplined and structured rules could be used while performing AD. One widely appreciated tools in the management of creativity and discussion in groups is Edward De Bono’s 6 thinking hats method (De Bono, 1989). Because of the way this tool is structured, it enables the entire team members to think in one direction at a given point of time. What happens in a 6 thinking hats session is that, a particular problem is discussed from one perspective at a given point of time. Because everybody on the team aims to discuss the problem in one direction, there is more collaboration among the team while one person is trying to react emotionally whereas another one reacting in a more rational way. It is perceived that this tool could be more useful during the AD sessions, as it promises the buy-ins of all the team members. So far, there has been no research studies on the impact of utilization of such methods in Affinity Diagram creation sessions and it is hence suggested that further research is performed to understand how it can improve the results and outputs of Affinity Diagram, to get a more cohesion and collaboration between the team members.

(46)

8 REFERENCES

Adams, M. E., Day, G. S., & Dougherty, D. (1998). Enhancing new product development performance: an organizational learning perspective.

Journal of Product innovation management, 15(5), 403-422.

Awasthi, A., & Chauhan, S. S. (2012). A hybrid approach integrating Affinity Diagram, AHP and fuzzy TOPSIS for sustainable city logistics planning.

Applied Mathematical Modelling, 36(2), 573-584.

Beyer, H., & Holtzblatt, K. (1999). Contextual design. interactions, 6(1), 32-42. Burchill, G., & Brodie, C. H. (1997). Voices into Choices: Acting on the Voice of the

Customer: Oriel Incorporated.

Burge, S. (2011). The systems thinking tool box

Cooper, R. G. (2001). Winning at New Products: Accelerating the Process from Idea to Launch (Создание успешных продуктов: ускорение процесса от возникновения идеи до выхода на рынок).

Cooper, R. G., & Kleinschmidt, E. J. (1993). Stage gate systems for new product success. Marketing Management, 1(4), 20-29.

Day, G. S., Schoemaker, P. J., & Gunther, R. E. (2004). Wharton on managing

emerging technologies: John Wiley & Sons.

De Bono, E. (1989). Six thinking hats: Taylor & Francis.

Elfvengren, K., Kärkkäinen, H., Torkkeli, M., & Tuominen, M. (2004). A GDSS based approach for the assessment of customer needs in industrial markets. International Journal of Production Economics, 89(3), 275-292. Fan, C. K., & Cheng, C. L. (2006). A study to identify the training needs of life

insurance sales representatives in Taiwan using the Delphi approach.

International Journal of Training and Development, 10(3), 212-226.

Foster, S. T., & Ganguly, K. K. (2007). Managing quality: Integrating the supply

chain: Pearson Prentice Hall Upper Saddle River, New Jersey.

Green, P. E., Carmone, F. J., & Fox, L. B. (1969). TELEVISION PROGRAMME

SIMILARITIES-APPLICATION OF SUBJECTIVE CLUSTERING. Journal of the

Market Research Society, 11(1), 70-90.

Griffin, A., & Hauser, J. R. (1993). The voice of the customer. Marketing science,

12(1), 1-27.

Gupta, A. K., & Wilemon, D. (1988). Why R&D resists using marketing information. Research-Technology Management, 31(6), 36-41.

Hauser, J. R., & Clausing, D. (1988). The house of quality. Harvard business review,

66(3).

Johnson, S. C. (1967). Hierarchical clustering schemes. Psychometrika, 32(3), 241-254.

Karypis, G., Han, E.-H., & Kumar, V. (1999). Chameleon: Hierarchical clustering using dynamic modeling. Computer, 32(8), 68-75.

King, R. (1987). Better Design in Half the Time: Implementing Quality Function Deployment (QFD) in America. GOAL, Lawrence, MA.

Kotler, P., Saliba, S. J., Turner, R. E., & Wrenn, B. (1995). Marketing Management:

Analysis, Planning, Implementation and Control, Canadian Eighth Edition, Philip Kotler, Ronald E. Turner. Instructor's Manual: Scarborough, Ont.:

(47)

Mazur, G. (1997). Voice of Customer Analysis: A Modern System of Front-End Tools. With Case Studies, AQC.

Minitab. (2016). What are the Goodman-Kruskal statistics? doi:dctm_Chron0900045780160138

Morgan, T., & Obal, M. (2016). Customer Participation in New Product

Development and New Product Performance: The Moderating Role of Expertise Celebrating America’s Pastimes: Baseball, Hot Dogs, Apple Pie

and Marketing? (pp. 465-465): Springer.

Mui, C. (2011). Five Dangerous Lessons to Learn From Steve Jobs.

Patnaik, D. (2009). Wired to care: How companies prosper when they create

widespread empathy: Ft Press.

Price, R. A., Wrigley, C., & Straker, K. (2015). Not just what they want, but why they want it: Traditional market research to deep customer insights.

Qualitative Market Research: An International Journal, 18(2), 230-248.

Rashid, N. R. (2012). A COMPARISON BETWEEN SINGLE LINKAGE AND COMPLETE LINKAGE IN AGGLOMERATIVE HIERARCHICAL CLUSTER ANALYSIS FOR IDENTIFYING TOURISTS SEGMENTS. IIUM Engineering

Journal, 12(6).

Renner, B. (2016). 2016 Consumer Products Industry Outlook | Deloitte US | Consumer Products. Retrieved from

http://www2.deloitte.com/us/en/pages/consumer-business/articles/consumer-products-industry-outlook.html

Rothwell, R. (1992). Successful industrial innovation: critical factors for the 1990s. R&d Management, 22(3), 221-240.

Sayad, S. (2016). Hierarchical Clustering. Retrieved from

http://www.saedsayad.com/clustering_hierarchical.htm

Shafer, S. M., Smith, H. J., & Linder, J. C. (2005). The power of business models.

Business horizons, 48(3), 199-207.

Shields, P. M., & Rangarajan, N. (2013). A playbook for research methods:

Integrating conceptual frameworks and project management: New Forums

Press.

Shillito, M. L. (2000). Acquiring, processing, and deploying: Voice of the customer: CRC Press.

Silverstein, D., Samuel, P., & DeCarlo, N. (2013). The innovator's toolkit: 50+

techniques for predictable and sustainable organic growth: John Wiley &

Sons.

Takai, S., & Ishii, K. (2010). A use of subjective clustering to support affinity diagram results in customer needs analysis. Concurrent Engineering. Ulwick, T. (2016). Customer Needs | Analysis and Assessment | Strategyn. Vanalli, S., & Cziulik, C. (2003). Seven Steps to the Voice of the Customer. Paper

presented at the DS 31: Proceedings of ICED 03, the 14th International Conference on Engineering Design, Stockholm.

Von Hippel, E. (1979). A customer-active paradigm for industrial product idea generation Industrial Innovation (pp. 82-110): Springer.

(48)

APPENDIX A : BRAINSTORMED QUESTIONS

No. Question to ask relevant

for whom comments

1

What goals do you try to achieve with your training programs? E.g. build competencies, entertain, create experiences, motivate? All of these? A mix of some?

2,3,4,5

2 How do you achieve those goals? 2,3,4,5

3

What sort of equipement and aids to you use to reach your training / workshop goals?

2,3,4,5 ask why for the things he / she

mentions

4

Have there been situations where you had difficulties reaching your training goals?

2,3,4,5

5 Why do you think you had those issues? 2,3,4,5

6 When would you use virtual aids and physical aids 2,3,4,5

7 What is the advantage of one over the other 2,3,4,5

8

What kind of trainings and workshops are your providing?

2,3,4,5

9

How many participants would on average attend one of your trainings?

2,3,4,5

10 What is the duration of your events? 2,3,4,5

11

What materials / means do you use to communicate your contents accross?

2,3,4,5

12

What kind of materials and equipment do you require during your training (such as. Flip charts and white

boards)?

How many of them do you require?

2,3,4,5 for each ask for what they require it for

13

Are there specific requirements for your trainings in terms of venue, space and setup?

2,3,4,5

14

Are specific issues you often run into when it comes to training equipment?

2,3,4,5 why are these issues not being

solved?

15

Are there any work arounds you need to use to make your trainings go the way you want to?

2,3,4,5 such as taping paper onto the wall because there is not enough space on whiteboards or flipcharts 16

Is there something that specifically annoys or bugs you when it comes to delivering your trainings (can

be participant behavior, venue, equipment, travel, anything)?

(49)

17

Can you compare the venue and equipment situation in China / India to other countries you have trained

you? What did you like more or less there?

2,3,4,5 are there specific reasons why the situation in China / India is like it is? 18

Was there something in terms of venue and equipment that once particularly impressed you?

2,3,4,5

19

If you had a magic wand and could create the perfect training environment for your trainings, what would

that look like? What kind of equipment would you have?

2,3,4,5 ask why

questions to dig deeper.

20

Are there any cultural influences that affect the way your training venues look like in the countries you

train?

2,3,4,5

21

How important do you believe training and workshop equipment to be for trainings and workshops as well

as for team collaboration in general?

2,3,4,5

22

What is your expectations out of a training program? Why do you attend a training?

14

23

Who normally decides which trainings you attend? You yourself? Somebody the company you work for?

14

24

how many trainings or workshops per year do you attend on average?

25 Who normally pays for the trainings you attend? 14

26

Is there a difference between the trainings you pay for and the trainings somebody else pays for?

14 according to

what criteria do you select which

training to attend

27

Describe the situation where you had really benefitted the training program

14

28

What sort of tools has the trainer in a typical training session that you know, uses?

14

29

What sort of equipment and aids (such as flip charts, whiteboards, etc.) are trainers of your trainings

using?

14

30

Describe the situation when you were annoyed/frustrated/not happy with the training aids

that the trainer had used

14

31

Describe a situation when the training equipment made a positive impact on the training and your

learnings

14

32

Describe a situation when you were at the highest of your creativity during a training? What caused that?

14

33

What impact do you think the training venue and training equipment has on the training and the

effectiveness of a training?

(50)

34

What sort of training aids and equipment (suach as flip charts and white boards) does your hotel provide

?

8.9

35 based on what criteria are these purchased? 8,9,13

36 who decides what material / equipment is purchased? 13

37

how do you know what materials and equipment to purchase?

8,9,13

38

How much do you usually allocate your budgets for such training aids

8,9,13

39 How does the storage of these equipments look like? 8,9,13

40

What roles does price play in your purchasing decisions? What are other criteria?

13

41

what kind of events happen in your hotel? How many of them are trainings / workshops?

8.9

42

what kind of equipment is normally requested for trainings and workshops? And how many?

8.9 this is important to ask, as I feel

that our trainings require more equipment than most of the other trainings that are held in

hotels. However, I am

not sure about this.

43

Is the material and equipment you have enough to fullfil the requests? Or do you run into situations where you cannot offer what the trainer or customer

wants?

8.9

44

do you get any kind of improvement feedback from customers towards your venue and the equipment in

particuluar?

8.9

45

How important is the quality of your venue and training equipment for your customers?

8.9

46

would high quality equipment be a positive differentiator for your hotel?

8.9

47

what are the main decision criteria according to which customers book a hotel as a training venue?

8.9

48

What do you consider while designing a training/conference/board room

11

49

What are the design criteria kept in mind while designing a room for the Trainers

11

50 How do the customers ask for what they want 12

51

What are the other doubts they would ask? For instance price?

12

52

what kind of trainings does your company organize, technical, soft skills

(51)

53

how many participants do you have on average per training

6

54

do you have your own training facilities or do you rent external facilities, or both?

6

55

are people also being sent to external trainings, or are all your trainings internally organized?

6

56

what is the character of the trainings you host? Is it mostly classroom / lecture style trainings with the teacher in front and participants listening or more

interactive trainings with group activities?

6.7

57

are there different venue and equipment requirements for these different types of trainings? If yes, are you

addressing this requirements?

6.7

58 Can you describe your training facilities 7

59 How are your training facilities equipped 7

60

what kind of feedback do you get towards your trainings facilities from trainers?

7

61

what kind of feedback do you get towards your trainings facilities from training participants?

7

62

according to what criteria do you select external training or workshop venues?

6.7

63

according to what criteria do you buy / pruchase trainign equipment such as flipcharts and

whiteboards

6.7

64

is there anything missing in terms of equipment that you would like to have?

6.7

65

are there any challenges, obstacles or limitations that trainers run into when using your training venues?

6.7

66

if you had all the decision making power, what would you change regarding your training venue?

6.7

67

have you been to an external training venue (maybe in another company, a hotel, abroad) that you member especially well? If yes, what made it

memorable for you?

6.7

68

what kind of trainings are you offering to your clients? In terms of topics and content

10

69

how would you describe the character of the trainings, is it more lecture style or interactive?

10

70

In terms of training character, what are your customers requesting?

10

71

Any trends you currently see in the training space in terms of topics but also training formats etc?

10

72

what kind of venue setup and equipment are your trainers requesting?

10

73

are the trainings held inside corporate venues or in external venues?

(52)

74

are your company customers able to fullfil the venue and equipment requirements?

10

75

have there been cases of very unusal training / workshop venue or equipment requirements? What

were these?

10

76

are you also requested to organize venue and equipment? If yes, what are your criteria for selecting

a venue and equipment?

10

77

how important is the equipment such as flipcharts, projectors, whiteboards for your overall work?

10

78

In terms of venue equipment, is there anything that would make your life easier or improve the quality of

trainings

10

ID

Types of people to interview / observe

(divergent list) why might we speak to them?

relevant for observations 14

Participants of a training Workshop

they know what is eventually needed for

them yes

2,3,4, 5

Local or foreign Trainers active in China

or India

they know their needs on site and have seen different training rooms and

facilities yes

6

Company employees responsible for training

planning

they should have an overview over what kinds of trainings and events are happening and maybe special needs

they have encountered yes

7

Company employees responsible for training

logistics

they know more details in terms of what

happens onsite yes

8.9

Hotel staff responsible for room bookings and

customers requests Hotel staff responsible

for room logistics

they have a broad overview over all

kinds of different requests in hotels yes

10 Trainer agencies

they have an overview over all kinds of different training suppliers, styles and

needs, as well as customer demands yes

13

Purchase Departments in Hotels / Purchase

Manager

they might know the budget constraints

How Subjective Clustering aids Affinity Diagram in grouping Customer needs in consumer products

How Subjective Clustering

aids Affinity Diagram in grouping Customer

needs in consumer products

SANDHEEP KUMAR VURUKKARA BOOPAL

How Subjective Clustering aids Affinity Diagram in

grouping Customer needs in consumer products

industry

Sandheep Kumar Vurukkara Boopal

Sammanfattning

Abstract

FOREWORD

NOMENCLATURE

Abbreviations

TABLE OF CONTENTS

1 INTRODUCTION

1.1 Background

1.2 Purpose

1.3 Method

1.4 Delimitations

2 FRAME OF REFERENCE

2.1 What are Customer needs and why should it be

structured?

2.2 Consumer products – What is different when it

comes to analysing customer needs

2.3 Affinity Diagram method

2.4 Subjective clustering method

2.6 Clustering

3 RESEARCH STUDY DESIGN

3.1 The innovation project studied

3.2 Brainstorming and filtering questions to get

relevant answers

3.3 Customer Interviews

3.4 Cleaning the data and deducing the list of

customer needs

3.5 Subjective Clustering and Affinity Diagram

3.6 Comparison and Analysis

4 RESULTS

4.1 List of customer needs

4.2 Comparison of grouped needs by AD and SC by

Product Development Team

4.3 Comparison of grouped needs by AD and SC by

Customers

4.4 Comparison of grouped needs of both AD and SC

by Product Development Team and Customers

4.5 Importance of needs

5 ANALYSIS AND DISCUSSION

5.1 Discussion

6 CONCLUSION

7 RECOMMENDATIONS AND FUTURE WORK

8 REFERENCES

APPENDIX A : BRAINSTORMED QUESTIONS