DEPARTMENT OF TECHNOLOGY MANAGEMENT AND ECONOMICS DIVISON OF SERVICE MANAGEMENT AND LOGISTICS CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2020 www.chalmers.se Report No. E2020:108 What is Quality in Coworking Spaces? Identification and Classification of Customer Needs Master’s thesis in Quality and Operations Management ERIK JOHANSSON ERIK RÅDMAN REPORT NO. E2020:108 What is Quality in Coworking Spaces? Identification and Classification of Customer Needs ERIK JOHANSSON ERIK RÅDMAN DF Department of Technology Management and Economics Division of Service Management and Logistic Chalmers University of Technology Gothenburg, Sweden 2020 A mixed method, multiple case study ERIK JOHANSSON ERIK RÅDMAN © ERIK JOHANSSON, 2020. © ERIK RÅDMAN, 2020. Supervisors: Petra Bosch-Sijtsema & Hendry Raharjo, Technology Management and Economics Examiner: Hendry Raharjo, Technology Management and Economics Report no. E2020:108 Department of Technology Management and Economics Division of Service Management and Logistics Chalmers University of Technology SE-412 96 Gothenburg Sweden Telephone +46 (0)31-772 1000 Cover: The three personas based on the identified customer segments, chapter 7. Typeset in LATEX, template by David Frisk Gothenburg, Sweden 2020 iv What Is Quality in Coworking Spaces? Identification and Classification of Customer Needs ERIK JOHANSSON ERIK RÅDMAN Department of Technology Management and Economics Chalmers University of Technology Summary Coworking spaces are receiving increased attention from scholars and property own- ers. This study aims to provide a first overview of customer needs in a coworking setting using a quality management perspective. A mixed method multiple case study approach was taken. A structured literature was carried out illuminating what coworking is, who coworks and motivations for use. Interviews and partici- pant observations were carried out at three coworking spaces in Gothenburg (Swe- den) highlighting 24 customer needs. A framework based on self-determination theory (SDT) was proposed to analyze qualitative findings. A web-based survey was conducted, designed and analyzed in accordance with the Kano model. The results provides three customer segments and three main recommendations in ac- cordance with Kano-classifications. First, it stresses the importance of a productive work environment and an ability to safely manage confidential information. Sec- ondly, coworking spaces can stand out amongst competitors with a high provision of front-end service and a professional work environment. The coworking customer is delighted by opportunities for collaboration, access to networks, knowledge sharing and wellness. The findings provide understanding of customer needs, particularly useful for owners and managers. The study lays a foundation for future research into customer needs in coworking spaces. Keywords: coworking, customer needs, quality management, Kano-model, self de- termination theory, affordance theory. v Acknowledgements First we would like to thank our supervisors Hendry Raharjo and Petra Bosch- Sijtsema who have spent numerous afternoons with us, discussing everything from research design to small details of the Kano questionnaire. Their help and commit- ment have made the past five months an enjoyable and productive journey. Secondly, we would like to thank the respective hosts Eva Nilson, Ebba Kullin and Carolina Wadsten who’ve always been warm and welcoming since our first day at their re- spective coworking spaces. They have introduced us to members and helped us with our requests throughout the whole research project, making it a smooth journey. We would also like to thank Pontus Wiking from United Spaces who, through numerous conversations over the phone and email, has helped us with the preparation and roll- out of our survey. A big thank you to all of the members we have interviewed, who have given us their time, insights, and valuable feedback. Furthermore we would like to thank Mia Johansson who has given us valuable insights into service design perspectives throughout the process. Erik Johansson & Erik Rådman, Gothenburg, May 2020 vii Contents List of Figures xiii List of Tables xv 1 Introduction 1 1.1 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1.1 What is missing? . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Research questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Scope & delimitation’s . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.5 Disposition of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 Theoretical framework 5 2.1 Quality management . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.1 What is quality? . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.2 Quality management theory . . . . . . . . . . . . . . . . . . . 6 2.1.3 The Kano model . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Psychological theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2.1 Self-determination theory . . . . . . . . . . . . . . . . . . . . 9 2.2.2 Affordance theory . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.3 Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3 Methods 13 3.1 Research design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.2 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.3 Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3.4 Qualitative methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3.4.1 Participant observation . . . . . . . . . . . . . . . . . . . . . . 16 3.4.2 Interviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.4.3 Documents and digital traces . . . . . . . . . . . . . . . . . . 19 3.4.4 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.4.5 Trustworthiness . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.4.6 Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.5 Quantitative methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 3.5.1 Kano-questionnaire . . . . . . . . . . . . . . . . . . . . . . . . 22 3.5.2 Response rate . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.5.3 Data cleaning and data analysis . . . . . . . . . . . . . . . . . 23 ix Contents 3.5.4 Validity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.5.5 Ethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4 Literature review 27 4.1 What is coworking? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.1.2 Typologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 4.1.3 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.1.4 Proposed benefits . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.2 Who is the user? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.3 Why cowork? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.4 What aspects are important? . . . . . . . . . . . . . . . . . . . . . . 36 4.5 Negative aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 5 Qualitative results & discussion 39 5.1 Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 5.2 Relatedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.2.1 To belong to a community (R1) . . . . . . . . . . . . . . . . . 42 5.2.2 To feel like a prioritized customer (R2) . . . . . . . . . . . . . 43 5.2.3 To feel welcomed at one’s workplace (R3) . . . . . . . . . . . . 44 5.2.4 That one’s workplace makes a good impression on guests (R4) 44 5.2.5 To have a workplace at a fixed location (R5) . . . . . . . . . . 45 5.3 Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.3.1 To know which other compaines are members and what they do (A1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 5.3.2 To be able to market one’s business and get exposure (A2) . . 46 5.3.3 To be able to choose a suitable work area depending on one’s current needs (A3) . . . . . . . . . . . . . . . . . . . . . . . . 47 5.3.4 To be transparent when meeting others without negative con- sequences for my business (A4) . . . . . . . . . . . . . . . . . 48 5.3.5 To be able to scale one’s business (A5) . . . . . . . . . . . . . 49 5.3.6 To have phone calls or conversations, without disturbing oth- ers (A6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 5.3.7 To be able to manage confidential information safely (A7) . . 51 5.3.8 To feel in control of social interactions (A8) . . . . . . . . . . 51 5.3.9 To be able to focus on work activities (A9) . . . . . . . . . . . 53 5.4 Competence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 5.4.1 To learn new things from peers and events (C1) . . . . . . . . 54 5.4.2 To be a healthy (C2) . . . . . . . . . . . . . . . . . . . . . . . 55 5.4.3 To meet people that can lead to business opportunities (C3) . 55 5.4.4 To cooperate/collaborate with relevant actors (C4) . . . . . . 56 5.4.5 To be able to receive help or input from others in a simple way (C5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 5.4.6 To have a workplace that gives you energy (C6) . . . . . . . . 57 5.4.7 To be able to work smoothly without disruptions (C7) . . . . 58 5.4.8 To be able to share my knowledge and competence (C8) . . . 59 5.4.9 To be able to focus on the core business (C9) . . . . . . . . . 59 x Contents 5.4.10 To feel an increased productivity from one’s workplace (C10) . 60 5.5 Summary of uncovered customer needs . . . . . . . . . . . . . . . . . 60 5.5.1 Relatedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 5.5.2 Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 5.5.3 Competence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 5.5.4 Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 6 Quantitative results & discussion 63 6.1 Kano results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 6.2 General discussion of Kano results . . . . . . . . . . . . . . . . . . . . 67 6.3 In-depth discussion of Kano results . . . . . . . . . . . . . . . . . . . 67 6.3.1 US - Segment 1 . . . . . . . . . . . . . . . . . . . . . . . . . . 69 6.3.2 US - Segment 2 . . . . . . . . . . . . . . . . . . . . . . . . . . 71 6.3.3 US - Segment 3 . . . . . . . . . . . . . . . . . . . . . . . . . . 72 6.3.4 A Working Lab . . . . . . . . . . . . . . . . . . . . . . . . . . 74 6.3.5 Comparing United Spaces and A Working Lab . . . . . . . . . 75 7 Synthesis and propositions 77 7.1 General recommendations . . . . . . . . . . . . . . . . . . . . . . . . 77 7.2 Personas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.2.1 Professional Penny . . . . . . . . . . . . . . . . . . . . . . . . 81 7.2.2 Networking Nicole . . . . . . . . . . . . . . . . . . . . . . . . 82 7.2.3 Efficient Edmund . . . . . . . . . . . . . . . . . . . . . . . . . 83 7.3 Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 7.3.1 Identified tensions and their implications . . . . . . . . . . . . 84 7.3.2 Implications of Kano-classifications . . . . . . . . . . . . . . . 84 7.3.3 Implications for future research . . . . . . . . . . . . . . . . . 85 8 Conclusion 87 References 89 A Interview template I B Photo elicitation V C Customer needs and Kano-questionnaire XI xi Contents xii List of Figures 2.1 The Kano model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2 Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.1 Flowchart of the structured literature review . . . . . . . . . . . . . . 15 4.1 Motivations to work at a coworking space . . . . . . . . . . . . . . . . 34 4.2 Why did you choose your current coworking space? . . . . . . . . . . 36 5.1 Uncovered customer needs in framework . . . . . . . . . . . . . . . . 40 6.1 Overall S/DS for United Spaces . . . . . . . . . . . . . . . . . . . . . 64 6.2 Gender of United Spaces respondents . . . . . . . . . . . . . . . . . . 68 6.3 Size of company of United Spaces respondents . . . . . . . . . . . . . 68 6.4 Type of membership of United Spaces respondents . . . . . . . . . . . 69 6.5 S/DS United Spaces segment 1 . . . . . . . . . . . . . . . . . . . . . 70 6.6 S/DS United Spaces segment 2 . . . . . . . . . . . . . . . . . . . . . 72 6.7 S/DS United Spaces segment 3 . . . . . . . . . . . . . . . . . . . . . 73 6.8 S/DS A Working Lab . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 7.1 Professional Penny . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 7.2 Networking Nicole . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 7.3 Efficient Edmund . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 xiii List of Figures xiv List of Tables 3.1 Interviewees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2 Wording of answers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.3 Original Kano evaluation table . . . . . . . . . . . . . . . . . . . . . . 24 3.4 Developed Kano evaluation table . . . . . . . . . . . . . . . . . . . . 25 4.1 Proposed benefits of coworking . . . . . . . . . . . . . . . . . . . . . 32 4.2 Motivations for coworking . . . . . . . . . . . . . . . . . . . . . . . . 35 5.1 Label and the corresponding need . . . . . . . . . . . . . . . . . . . . 41 6.1 Abbreviations reminder . . . . . . . . . . . . . . . . . . . . . . . . . . 63 6.2 United Spaces results . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 6.3 United Spaces survey response . . . . . . . . . . . . . . . . . . . . . . 65 6.4 A Working Lab survey response . . . . . . . . . . . . . . . . . . . . . 66 6.5 United Spaces segment 1 . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.6 United Spaces segment 2 . . . . . . . . . . . . . . . . . . . . . . . . . 71 6.7 United Spaces segment 3 . . . . . . . . . . . . . . . . . . . . . . . . . 73 6.8 A Working Lab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 7.1 Managerial Implications . . . . . . . . . . . . . . . . . . . . . . . . . 79 7.2 Relative importance of uncovered customer needs . . . . . . . . . . . 80 C.1 Customer needs and Kano questions . . . . . . . . . . . . . . . . . . XII xv List of Tables xvi 1 Introduction 1.1 Problem statement Coworking spaces is a relatively new service and a serious contender to the traditional office as the setting in which work takes place. The number of coworking spaces has increased worldwide from 8’900 in 2015 to 18’700 in 2018 and the number of members has increased by 300 percent during the same period (Deskmag, 2019). The Covid 19 pandemic, has exposed a drawback with normal offices on an unprecedented scale: Why do companies need all this space? Johns and Gratton (2013) uses a curious analogy trying to capture the essence of coworking. Coworking spaces are to knowledge work what bike-share programs are to transportation: a community-based, low-cost, convenient, and eco- friendly solution.(p.71) In short a coworking space is a shared office where workers from various fields and companies are co-located through a membership fee. Since the beginning of coworking in San Francisco in 2005 (Gandini, 2015) there has been a number of notable undertakings to understand the phenomenon (Bilandzic & Foth, 2013; Capdevila, 2013; Gandini, 2015). Many benefits are proposed in research such as providing a sense of community (Parrino, 2015), collaboration (Capdevila, 2014), innovation (Yang, Bisson, & Sanborn, 2019) and productivity (Salvador, Gon- zalo, & Dolores, 2018). Several studies are also conducted relating to the users of coworking spaces and their motivations for using the service. 1.1.1 What is missing? Although the body of research has been growing, there is still an ambiguity and confusion regarding many aspects of the phenomenon (Gandini, 2015; Clifton, Füzi, & Loudon, 2019) and few quantitative studies have emerged. Given the many efforts made to understand the coworking users and their motivations it is surprising that little attention has been directed to customer needs. There is a growing interest from providers and property owners to understand their customers and how to cater to their needs. Few scientific inquiries have been undertaken from the field of quality manage- ment. A key aspect of modern quality management theory is its focus on customers 1 1. Introduction (Bergman & Klefsjö, 2010), yet focusing on the customers is by no means an easy task. The number of potential customers are finite; their attention and money is not easily won. It is of great importance for a business to uncover and classify customer needs and direct its resources to where it matters for the customer. 1.2 Aim Firstly, the aim is to review and structure the extant literature about coworking. More specifically to understand what coworking is, who coworks, and reasons for using coworking. Secondly, the aim is to uncover and classify customer needs in coworking. 1.3 Research questions We intend to reach our aim through trying to answer the following research ques- tions: 1. What is the current state of coworking literature? 2. What are the needs of the coworking customer? 3. How can these needs be classified according to the Kano model? 1.4 Scope & delimitation’s The review of the state of the coworking literature is done at a specific point in time and will reflect the research at that particular moment. The scope of the customer and user perspective will primarily be that of the quality management field, with a helping hand from self-determination theory and affordance theory. Several methods of classifying customer needs exist within the Quality Management field, yet our focus will be on the Kano model. Furthermore the study is situated at three coworking spaces in the city of Gothenburg, Sweden, covering the demographics of those coworking spaces. Moreover, business models and ecological implications of coworking is not within the scope of the study. 1.5 Disposition of the thesis 1. Introduction Sets the scene and introduces the reader to the research questions and what they aim to answer 2. Theoretical Framework Describes the theoretical background that is used throughout the study. In- troduces the reader to the theoretical framework to be used in the analysis. 2 1. Introduction 3. Methods Describes the research approach, data collection and analysis. Also discusses the validity and ethics of the research. 4. Literature Review Summarises the coworking literature, specifically regarding what coworking is, insights into motivations for coworking and who the coworking customer is. 5. Qualitative Results & Discussion Highlights the qualitative findings using a proposed framework. 6. Quantitative Results & Discussion Highlights the quantitative results using the Kano-model and S/DS-diagrams. The responses from United Spaces is treated as a whole and as three separate customer segments. A Working Lab is treated as its own segment. 7. Synthesis and propositions The qualitative and quantitative findings are discussed together. Three per- sonas based on the identified customer segments are presented. Some identified tensions and implications are outlined as well. 8. Conclusion Concluding remarks that portrays our research contribution and key take- aways. 3 1. Introduction 4 2 Theoretical framework The following chapter introduces the reader to the theory that will be used in de- velopment of methods and in discussion of findings. 2.1 Quality management The concept of quality and what it is plays a central role in the quality management field. We will begin by explaining what quality entails, historical definitions and more recent ones. Then, quality management theory is explained, followed by the Kano model. 2.1.1 What is quality? The concept of quality has been around for several millennia: from building the pyramids of Egypt to rifle manufacturing in the 19th century (Bergman & Klefsjö, 2010; Juran, 1951). Walter A. Shewart is by many seen as one of the fathers of the quality management field. In his book "Economic Control of Quality of Manufactured Product", Shewhart (1931) define quality as follows: ...there are two common aspects of quality. One of these has to do with the consideration of the quality of a thing as an objective reality indepen- dent of the existence of man. The other has to do with what we think, feel or sense as a result of the objective reality. In other words, there is a subjective side of quality. (Shewhart, 1931, p. 53) Furthermore, Shewhart (1931) explains that the goal of a production system is to cater for the needs of humans, the customers. It is the customer who judges the subjective qualities of a product or service. Crosby (1980) uses a more compact definition of quality: All through this book, whenever you see the word ’quality,’ read ’confor- mance to requirements’. (Crosby, 1980, p. 15) He thereby leaves out the subjective part of quality in his definition, though he still emphasizes the importance of satisfying the customer needs throughout his writings (Crosby, 1980). The customers judgement of quality has always played a central role in the quality management field, as Deming (1982) put it: 5 2. Theoretical framework The customer is the most important part of the production line. Qual- ity should be aimed at the needs of the customer, present and future. (Deming, 1982, p. 4) The customer is the raison d’être for the whole organization, and without them there would be no one to purchase the product or service. The importance of the customer relationship should not be underestimated. The definition of quality made by Harry, Mann, De Hodgins, Hulbert, and Lacke (2011) entails this symbiotic relationship between organization and customer: "Quality is a state in which value entitlement is realized for the customer and provider in every aspect of the relationship". (Harry et al., 2011, p. 5) All of the aforementioned definitions are explaining what quality is. Taguchi and Wu (1979) do the opposite, and define quality by what it is not: The lack of quality is the losses a product imparts to the society from the time the product is shipped (Taguchi & Wu, 1979, p. 76) As has been noted above, the concept of quality and definition of the term has been developed over the years. The definition that will be used throughout the thesis is that by Bergman and Klefsjö (2010) where they emphasise the aspect of exceeding what is expected by the customers: The quality of a product [article or service] is its ability to satisfy or preferably exceed the needs and expectations of the customers. (Bergman & Klefsjö, 2010, p. 23) 2.1.2 Quality management theory As with most management theories since Taylor (1911), one of the key proposed benefits of quality management is an increased productivity. Bergman and Klefsjö (2010) mention several other promises of quality management such as: • more satisfied and loyal customers • lower employee turnover and sick leave rates • a stronger market position • shorter lead times • opportunities for capital release • reduced waste and rework (Bergman & Klefsjö, 2010, p. 54) Quality management aims at fulfilling these promises by utilizing a set of values, techniques and tools (Hellsten & Klefsjö, 2000). Bergman and Klefsjö (2010) call these set of values guiding principles which are the foundation, or the cornerstone of the quality management field. The central role of focusing on one’s customers is a key aspect in quality management theory (Bergman & Klefsjö, 2010; Dean Jr & Bowen, 1994; Grönroos, 1984). Yet, focusing on the customers is by no means an easy task. The number of potential customers are finite, and their attention and money is not easily won in today’s global competitive market. How does one 6 2. Theoretical framework find out what existing and potential customers need in order to develop product or service attributes that cater to those needs? By identifying the so called voice of the customer, a service or product provider can learn what the customer needs. Bergman and Klefsjö (2010) highlights the difficulties in doing just that: Focusing on customers implies finding out what they want and need, and then to systematically try to fulfil these needs and expectations when developing and manufacturing the product. It is not always easy to as- certain what the customers want, using, for example, market surveys. often the customers themselves are not able to state their needs, and it takes considerable empathy to understand what they really need. (Bergman & Klefsjö, 2010, p. 38) Several methods exist within the Quality Management field, and one way to tackle this challenge is to use the methods of Noriaki Kano. 2.1.3 The Kano model The Kano model was introduced in the paper "Attractive quality and must-be qual- ity" by Kano (1984). The model has been used extensively within the quality man- agement field. Kano (1984) argue that there are different types of customer require- ments which can be grouped into distinct categories. Each category contributes in a different way to the level of customer satisfaction (Kano, 1984). The original categories are the following: 1. Attractive An attractive quality is one that while present leads to customer satisfaction, but won’t lead to dissatisfaction if not present in the product or service. 2. One-dimensional For one-dimensional product qualities, the level of satisfaction is proportional to the level of provision. A higher level of provision translates to higher sat- isfaction. A poor provision of one-dimensional qualities leads to customer dissatisfaction. 3. Must-be Must-be qualities are those that when left unprovided leads to customer dis- satisfaction, but when provided does not lead to satisfaction. 4. Indifferent A customer can be indifferent towards an attribute which will not affect the overall impression of the product or service. An indifferent quality corresponds to the horizontal line in Figure 2.1. (Berger, Blauth, & Boger, 1993) 7 2. Theoretical framework Low High High Provision One-dimensional Must-be Attractive Low Customer satisfaction Indifferent Figure 2.1 The Kano model, adapted from Berger et al. (1993) There is also the possibility of a reverse quality where a high provision leads to low satisfaction. This can be the case when a company has mistaken what their customers need (Berger et al., 1993). What type of product or service attribute that belongs to which category differs between customers. A distinct cluster of customers with similar sentiments to the categories can be thought of as a market segment (Berger et al., 1993). Kano (1984) developed a questionnaire with the purpose of categorizing different product and service attributes into the different categories. In a Kano questionnaire the questions are asked in pairs, the first in the pair is called the functional form (F) and asks about how the customer would feel if an attribute were to be provided. The second in the pair is called the dysfunctional form (D) and asks how the customer would feel if the same attribute were not provided. Each pairwise answer will then be categorized into one of the categories of the Kano model. Matzler, Hinterhuber, Bailom, and Sauerwein (1996) proposes a way to measure to what extent a product- or service attribute leads to satisfaction or dissatisfaction. To get an overview of the importance of an attribute, Matzler et al. (1996) proposes a figure where satisfaction is plotted on the y-axis and dissatisfaction on the x-axis. Satisfaction and dissatisfaction is measured using the relative importance of each Kano category from the questionnaire. This provides an understanding of variation regarding customer preferences. The value for satisfaction (S) goes from zero to one while the dissatisfaction (DS) value goes from zero to minus one and are calculated 8 2. Theoretical framework using the following formulas: S = A + O A + O + M + I (2.1) DS = − O + M A + O + M + I (2.2) The letters in the formulas above correspond to the number of respondents vieweing a specific quality as attractive (A), one-dimensional (O), must-be (M) or indifferent (I). An important aspect of Kano’s theories is that customer sentiment towards a cat- egory changes over time (Berger et al., 1993). Attractive qualities become one- dimensional, one-dimensional qualities become must-be qualities. One example be- ing air conditioning in cars that when introduced was seen as an attractive quality. But today most people have come to expect air conditioning in their cars, meaning it has gone from an attractive to a must-be quality. As mentioned by Berger et al. (1993), the Kano model is mostly used to measure the importance of product or service features, but it can also be used to directly measure customer needs. A fea- ture or affordance corresponding to a certain need can have a different classification than the need itself. Different features and affordances can contribute differently to the corresponding need. Since its introduction the Kano model has been used in academia and by practitioners in marketing, product development and research & development to create and deliver product and service offerings that are aligned with customer needs. 2.2 Psychological theory Up to this point many studies have focused on why people choose to cowork, pro- viding findings on motivations tied to specific settings. However, as time changes, so does people’s motivations. Therefore it is relevant to ground motives in deeper psychological needs. Self-determination theory will be introduced as a means to un- derstand the underlying needs of the coworking customers. Thereafter, affordance theory is introduced as a way to understand how coworking can realise these needs and to formulate the questions for the Kano-questionnaire. 2.2.1 Self-determination theory Self-determination theory (SDT) was introduced by E. L. Deci and Ryan (1985). There are three basic psychological needs that form the basis of intrinsic motivation in humans: autonomy, competence, and relatedness (Ryan & Deci, 2017). The theory suggests when these needs are fulfilled results in intrinsic motivation and leads to thriving for individuals, psychological growth, general well-being and content (E. L. Deci & Ryan, 1985, 2000). When unfulfilled, it causes psychological distress and lack of well being. Furthermore these needs are universal and objective to humans, as Ryan and Deci (2017) puts it: 9 2. Theoretical framework This assertion is analogous that whether or not one subjectively val- ues, desires, or prefers vitamin C, extended deprivation of it will lead to scurvy (Ryan & Deci, 2017, p. 10) The theories of Deci and Ryan have been, and still are, applied in numerous areas. One area in which it has been proven useful is in management and work place moti- vation by organizational psychologists (E. L. Deci & Ryan, 1985; E. Deci, Olafsen, & Ryan, 2017). ...all human beings have three fundamental psychological needs—for com- petence, autonomy, and relatedness—which when satisfied promote au- tonomous motivation, wellness, and effective performance. Thus, SDT has been centrally concerned with promoting the need-supportive con- ditions across domains that facilitate people motivating themselves au- tonomously and in turn working well and feeling good. (E. Deci et al., 2017, p.39) Relatedness The need for relatedness is about the reciprocal caring and concern between indi- viduals and groups of people (Ryan & Deci, 2017; E. L. Deci & Ryan, 2000); "To love and care, and to be loved and cared for" (E. L. Deci & Ryan, 2000, p. 231). There can be trade-offs between the need for relatedness and the need for autonomy. Relatedness within a group or between individuals usually triumphs at the expense of a person’s need for autonomy (E. L. Deci & Ryan, 2000). Autonomy Autonomy is the need to be in control of one’s situation and actions (Ryan & Deci, 2017). The need for autonomy is deeply ingrained in human beings and creates a tendency to take actions that are aligned with one’s personal goals and needs. To feel as if one’s actions are the results of free will, and not forced upon oneself by others or the current situation (E. L. Deci & Ryan, 1985, 2000). Furthermore E. Deci et al. (2017) explain that in a workplace context "...when employees have a sense of autonomy they themselves find ways to get the other needs satisfied" (p. 23). Competence Competence is the need to use one’s skills, knowledge, and overall capacity to inter- act with the physical and social environment in the most effective way (E. L. Deci & Ryan, 1985, 2000). The need for competence creates an urge to develop one’s skills and apply them on tasks that are challenging but possible to complete. As a person’s skills and knowledge develops, he or she will seek out new challenges in order to further improve those skills (E. L. Deci & Ryan, 2000). "Where individuals are prevented from developing skills, understanding, or mastery, the competence need will be unmet." (Ryan & Deci, 2017, p. 86) 10 2. Theoretical framework 2.2.2 Affordance theory The concept of affordances has its origin in the field of Ecological Psychology and was introduced in the 1960s by the psychologist James J. Gibson. In his book "The Ecological Approach to Visual Perception" from 1979, Gibson made the following definition of affordances: The affordances of the environment are what it offers the animal, what it provides or furnishes, either for good or ill. (Gibson, 1979, p. 127) Gibson (1979) states that the perception of affordances are dependent on culture, previous experience, and needs. A hungry person would perceive the affordances of an apple differently than he or she would had that person already eaten. The percep- tion is different but the affordance "to be eaten" is there in both cases. Furthermore, Gibson (1979) argues that the perception of an object or an environments affordance is unique to each organism, although at many times a set of affordances of an object is perceived similarly by many, but not all. For example, many would agree that one of the affordances of a pen is to write, but a person not knowing about the concept of writing would not be able perceive the same affordances (Gibson, 1979). Affordance theory has been used in numerous fields since its inception, most notably in product- and interaction design, where Norman (1988) introduced the concept in his work "The Design Of Everyday Things". Norman (1988) means that an af- fordance is a property of the object or environment, whereas Gibson (1979) argued that an affordance is a relational property between the organism and the object or environment. Gaver (1991) highlights the usefulness of affordance theory when he applies it in the context of understanding user behaviour in interface design. Gaver (1991) uses the following definition of affordances: Affordances are properties of the world that are compatible with and rel- evant for people’s interactions (Gaver, 1991, p. 79). Gaver (1996) uses the concept of affordances to explore the role the built environ- ment plays in the interaction between individuals: Interior decoration relies on designing affordances: The ways tables are laid out in restaurants, for instance - whether they are small and well spaced, or arranged in long rows - will determine whether the space offers an intimate encounter or a convivial celebration. (Gaver, 1996, p. 121) Furthermore Gaver (1996) argues that the perception of "affordances for social in- teraction" can differ between societies, which is in line with the cultural aspect of affordances proposed by Gibson (1979). Affordance theory is to this day used as a way to understand how humans interact with the environment and newer phenomena such as the use of social media platforms has been viewed through its lens (Bucher & Helmond, 2017). Karahanna, Xu, Xu, and Zhang (2018) combine affordance theory with SDT and psychological ownership theory to create a lens thorough which they view social media: 11 2. Theoretical framework The paper develops a needs–affordances–features (NAF) perspective on social media use which posits that individuals’ psychological needs moti- vate their use of social media applications to the extent to which these applications provide affordances that satisfy these needs. (Karahanna et al., 2018, p. 737) 2.3 Framework The following framework has been developed from SDT. The identified needs will be placed in the most closely related SDT dimension. Kano-classifications are indicated with colors. Relatedness Autonomy CompetenceAttractive One-dimensional Must-be Indifferent Kano classification Figure 2.2 Framework 12 3 Methods The study took place at three different coworking spaces in the city of Gothenburg, the second largest city of Sweden. The researchers were granted access as members of the coworking spaces and spent over 900 hours immersed. During the course of the project, the research team had weekly meetings to discuss the observations and findings, and to guide further data collection. 3.1 Research design Choices regarding research design and methods, need to harmonize with the research questions (Bryman & Bell, 2011). The first research question is a prerequisite for the second, meaning that the initial inquiry becomes to find what customer needs exists in coworking. Qualitative studies are a first step in quantitative studies where relevant factors are identified to later be measured (Wallén, 1996). Since very few studies have been made with inquiry regarding customer needs in coworking a qual- itative study need to take place in order to find these needs. These needs can then be measured quantitatively. Uncovering customer needs entails understanding not only explicit aspects of the coworking phenomenon, but implicit one’s as well such as members subjective perspectives and various social aspects such as relations be- tween actors. Wallén (1996) argues that qualitative studies are necessary for things that are vague, ambiguous and subjective like experiences and feelings that are hard to quantify. In this sense our inquiry is of a quite exploratory character in that we try to view the coworking phenomenon through a new lens, through interviews and observations, and build theory through generalizations made from observations, in line with an inductive approach (Bryman & Bell, 2011; Wallén, 1996). Wallén (1996) proposes the case study as a method enabling one to see what hap- pens under real circumstances and from these observations generate deep knowledge about these events. The case study approach is also a common avenue for scientific inquiry in various fields, including management research (Yin, 2009). Customer needs cannot be isolated from the product or service they relate to, rather these needs are instead often intimately tied to the related product or service. Therefore a case study approach is needed and it also works well with coworking since these spaces easily can be identified as demarcated systems (Merriam, 1994). To be able to contrast findings and not be to dependent on one case, a multiple case study approach is used. 13 3. Methods No research takes place in a vacuum. Even though very few studies had explic- itly looked at customer needs in coworking many inquiries have been done in the phenomenon. Therefore a structured literature review was done in order to get a picture of the contributions in the field, specifically relating to what coworking is, who uses this service and why these people cowork. Yin (2009) views thorough literature reviews as a natural starting point when employing a case study method. This theory was used in discussion in order to see whether the identified needs could be verified or contrasted to prior research as well as if our endeavor had uncovered new aspects or missed existing ones. Yin (2009) argues case studies follow linear but iterative path and Bryman and Bell (2011) also notes that many inductive studies have iterative patterns of going back and forth between theory and data. Some scholars reshape this view slightly and refers to this process as abductive (Dubois Gadde, 2002). As our method below will clarify our initial qualitative endeavor was to a high extent inductive. To test the generated theory and to be able to answer the second research question a quantitative study took place in the form of a self-completion questionnaire. This part is deductive since the uncovered customer needs are "subjected to empirical scrutiny" (Bryman & Bell, 2011, p. 11). In this sense our research design can be said to have to follow a mixed methods, multiple case study research approach. An inductive endeavor builds theory about customer needs from findings. This theory is thereafter tested quantitatively. The data is gathered from multiple coworking spaces in both instances. 3.2 Literature review A structured literature was initialized by searching on web of science using the keywords: coworking, cowork, co-working, co-work, co working, co work. In the first screening the identified literature was screened for articles that were about coworking. For the second step, articles with an SJR of one or two were kept and the content of the articles were examined more closely. It consisted of reading the abstracts and conclusions in order to identify if the article answered who coworks what coworking is, and why people cowork. Before the third and last screening, as suggested by (Randolph, 2009), a snowball search from several was performed as a way to find more relevant literature and reach saturation. The third and last screening was done according to the judgement of the researchers.The literature review took place between February 6, 2020 and February 17, 2020. An overview of the process is illustrated in figure 3.1. 3.3 Setting United Spaces has over 1’300 members in Sweden. They opened in Gothenburg in 2017 and is located in the city centre with roughly 300 members. Anyone is allowed 14 3. Methods Articles found in database searches 164 97 72 63 67 First screening: coworking 40 Second screening: SJR and “who, what or why” 9 Third screening: researchers judgement 15Articles found from snowballing Figure 3.1 Flowchart of the structured literature review to join and become a member of the coworking space and partake in their activi- ties, such as seminars, breakfasts, members lunch. Three different memberships are available: a lounge membership which grants access to an area similar to a hotel lobby, a flex membership giving access to a area with ergonomic chairs and desks and private offices, which can be customized to suit tenant’s needs. Amenities and service functions are included in all three memberships. A Working Lab opened during late fall of 2019 and is located outside the city centre in a science hub dedicated to civil engineering and urban planning. A Working Lab’s approach is to function as an interface between industry and academia. A Working Lab is selective regarding members in order to ensure that users fit their intended theme. A Working Lab has two types of memberships, a flex membership which gives access to an area with various work spaces where the member chooses a work area dependent on the current work needs. The other type of membership are private offices which are part of the same area as the flex membership. The offices differ in size to suit different company sizes. Members with a private office have access to the open area in the same way those with a flex membership have. Arena Nordstan opened in march of 2020 located in the city centre. The aim with this space is to function as a pilot test for future coworking spaces. Arena Nordstan is owned by the property owner Vasakronan and the idea is to offer coworking memberships to Vasakronans existing tenants. Arena Nordstan has the same type of memberships as A Working Lab. All three coworking spaces are owned by large property owners. The similarity in ownership could have some implication on how representative the findings are relating to the coworking spaces phenomenon as a whole. This might be remedied 15 3. Methods somewhat by the spaces different focus and since the three spaces were in different stages of operation (United Spaces experienced, A Working Lab and Arena Nordstan newly opened), it gave access to individuals with different amount of experience of coworking spaces. 3.4 Qualitative methods 3.4.1 Participant observation One part of understanding the needs of someone else is to try to see things from this persons perspective. Given our aim to uncover customer needs in coworking, participant observation was therefore one chosen method. Participant observation can be defined as: ...a way to collect data in naturalistic settings by ethnographers who ob- serve and/or take part in the common and uncommon activities of the people being studied. (Dewalt Dewalt 2011, p.2) The method has often been used in studies situated in organizational settings (Bryman & Bell, 2011). Clancey (2006) argues that being able to actually un- derstand work practice and inherent conflicts of work settings can be hard for an outsider, yet these tensions creates constraints. These constraints can also be a promising source of information in a process to uncover customer needs, when not only straightforward aspects are of interest. DeWalt and DeWalt (2011) argues that the method can help to form a holistic and deeper understanding of phenomenon. It can also give researchers an intuition about how people relate to one another, facilitate the research process as well as provide experiences that can be discussed with other members (Schensul & LeCompte, 2012). Clancey (2006) also mentions visual analysis and its ability to appreciate more straightforward aspects such as pos- tures and gestures, but also more elaborate things such as preferences for privacy or indirect involvement. During a period of six months, the researchers spent their time as members at the three coworking spaces, performing participant observations. Both researchers were present most days of the week, usually through normal office hours. Occasionally the researchers were there during early mornings, late evenings and weekends to contrast previous experiences. Similarly an effort was made to experience as many rooms and seating areas of the settings as possible. The researchers attended events such as weekly member breakfasts, physical and remote seminars and remote workouts. This resulted in over 900 hours between the two spent researchers at coworking spaces during the time of research. Guba, Lincoln, and others (1994) argues that trustworthiness of a study increases if considerable time is spent in the setting. An overt ethnographic approach was taken. At A Working Lab and United Spaces community managers introduced the researchers and the research to other members when given the opportunity, thereby providing access and legitimacy of inquiry well in line with Clancey (2006). At A Working Lab the information about the 16 3. Methods researchers presence was also sent out in a member slack channel. Field notes were taken in a project diary. Novel experiences, thoughts, and obser- vations were written down and recorded with setting, date and time following the suggestions of Clancey (2006) and Schensul and LeCompte (2012). Furthermore an effort was made to keep notes descriptive, separating own thought or experiences from observations. An important aspect made possible by observation was experiencing the settings by working there as members. The role of the researchers could be situated somewhere in between participant as observer and observer as participants according to Gold’s (1958) nomenclature, since they were full members, yet with a focus on data collec- tion and understanding user needs. The ethnographic approach allowed not only for direct findings, but for a greater understanding of interviewees and their accounts. One limitation of the ethnographic inquiry might be that both researchers were male possibly influencing accessibility to some avenues of information (DeWalt & DeWalt, 2011). 3.4.2 Interviews Thirteen semi-structured member interviews were held throughout the study. The main purpose of the interviews was to identify the customer needs which would serve as the input for a Kano questionnaire. A purposive sampling was done in order to cover a diverse group of interviewees with respect to some characteristics (Bryman & Bell, 2011) such as age, gender, occupation, employment type and time as a members. Several of the interviewees had previous experiences from other coworking spaces or cafes which enabled them to contrast their experiences. Table 3.1 Interviewees m=male, f=female Code CS Age Profession Employment M001 Arena 25-34 Sales Manager Permanent M002 US 25-34 Entrepreneur - Business Developer Self Employed M003 Arena 25-34 Corporate Development Manager Permanent F004 AWL 35-44 Business Developer - HR Permanent M005 AWL 25-34 Data Engineer Permanent F006 US 25-34 Solution Engineer Permanent F007 AWL 35-44 Entrepreneur - Process Leader Self Employed M008 US 45-54 Manager - Business Developer Permanent M009 US 45-54 Entrepreneur - Business Developer Self Employed M010 US 55-64 Entrepreneur - Business Developer Self Employed F011 US 25-34 Management Consultant Permanent F012 US 55-64 Entrepreneur - HR Self Employed F013 US 45-54 Regional Manager Permanent Most interviews were partaken by both researchers, making it possible for one to 17 3. Methods focus more on being present in the conversation, whereas the other could focus on taking notes and sometimes intervene with follow-up questions. This way specific emotions could be captured in the moment and used as specific points of interest during transcription. In the beginning of each interview the interviewee was asked to map his or her energy levels throughout their day and explain why it rose or fell (Appendix A). The purpose with this exercise was to get a basic understanding for how a normal day looks like in their life. The mapping was removed from later interviews due to saturation and in some cases time-constraints. Photo-elicitation was used as a way to uncover more information than with regular interview-questions alone (Harper, 2002; Clark-IbáÑez, 2004). The interviewees were asked to pick three pictures out of 22 that they associated with something that made their work easier by being at a coworking space (Appendix B). They were also asked to chose one picture that they associated with something frustrating regarding coworking. By asking what’s frustrating about one’s job or task, potential product or service solutions can be uncovered (Ulwick & Bettencourt, 2008). The photo elicitation combined with follow-up questions proved to be the most useful method to uncover information related to our research. After the photo-elicitation the rest of the interview was done in a semi-structured manner. In semi-structured interviews questions are of more general character com- pared with structured ones and the researcher has room to pursue new questions sparked from replies (Bryman & Bell, 2011). At the beginning of the study this part had a more exploratory character in the sense that interviewees were asked to elaborate on many aspects. Interviewees were allowed to take off on tangents providing indications of aspects of importance to the interviewee (Bryman & Bell, 2011). Later when some aspects had been confirmed multiple times, less follow-up questions were asked, instead focus shifted to new information. Follow-up questions were often just asked to make the interviewee clarify certain points or aspects, but sometimes also to ask how often something occurred. Sometimes during interviews the interviewees would get more emotional regarding a certain issue. It turned out that findings from these moments often could be coupled with a need. A few interviews were carried out with video conference solutions as a result of the Covid-19 pandemic. The journey mapping was excluded from these interviews, but the photo exercise was kept, through digitally sending a file with the photos to the interviewees. The video interviews worked well and the researchers did not perceive any barriers in extracting information. All interviews were recorded and transcribed. This proved to be a good practice enabling repeated examination of the material and rigorous analysis (Bryman & Bell, 2011). A fourteenth interview was held with a member from a fourth coworking space “Entreprenörsgatan”, in order to contrast previous findings. More interviews were intended to be held with members at this place, but this became difficult due to the Covid-19 pandemic. At this point saturation was already reached at the other 18 3. Methods coworking spaces. During one of the weekly meetings with the research group, a discussion arose related to what kind of social theory could be appropriate discussing the findings. What could explain how people use and interact in these spaces? Self-determination theory (SDT) came up as a potential candidate and it was de- cided to incorporate some interview questions relating to SDT. Questions previously used for this purpose (Van den Broeck, Vansteenkiste, De Witte, Soenens, & Lens, 2010; E. L. Deci & Ryan, 1985, 2000) were slightly tailored to a coworking setting. It turned out that the answers showed to be consistent in the sense that no con- tradictions were found between negative and positive statements. Some questions did however only result in yes or no answers in some cases. Perhaps more interest- ingly the answers were not always relating to the aspect of SDT being asked about. The questions however showed to provide new insights on customer needs and were therefore kept and developed to some degree. 3.4.3 Documents and digital traces Documents and digital traces are not produced as part of research inquiry, yet can be used by researchers in analysis of phenomenon (Bryman & Bell, 2011). The researchers naturally had access to the websites of the coworking spaces, providing material regarding history, membership plans, prices and so forth. Beside these straightforward sources some additional ones were encountered. These data gave an additional lens of information for the process of uncovering customer needs. United Spaces At United Spaces a member survey was made accessible for the researchers. This survey was done by a third party providing United Spaces with net promoter scores on various features of the service. It also gave members the option to freely write reactions as well as desires connected to the service. As all members at United Spaces the researchers got access to an internal mobile application where one can connect with others, book meeting rooms and purchase various products and services. Due to the Covid-19 pandemic United Spaces organized various digital events for their members. Some of these were attended by the researchers, such as seminars and remote workouts led by an instructor. These events also featured a chat-function for members to use. A Working Lab Slack is used as an internal communication channel for members of A Working Lab. The researchers were included in this channel. As with United Spaces, A Working Lab hosted digital events following the Covid-19 pandemic. This took the form of a company presentation done by one of the members with a following round of questions. 19 3. Methods 3.4.4 Data analysis The field notes and diary were discussed during the weekly research group meet- ings to highlight, theorize, and contrast findings with existing literature, our own experiences and observations. 20 3. Methods Observations were coded into nodes in NVivo to give an overview of what had been observed and experienced. Similarly all of the interviews were coded in NVivo. The nodes from the observations and interviews were grouped in similar parent nodes and themes. Different types of categorization were tried in order to reach new insights. This was done in several iterations as more themes emerged. Themes were grouped according to similarities and higher categories emerged. PowerPoint was used as a tool for visual presentation and development of the analysis of findings from observations and interviews. Articles included in the literature review were coded in NVivo to create an understanding for common themes and findings relevant to the research purpose. 3.4.5 Trustworthiness Credibility is mainly achieved through triangulation, namely the study of various sources of information to uncover customer needs (Bryman & Bell, 2011). The combination of interviews and participant observation proved specifically useful; Own experiences could be confirmed by other’s accounts and vice versa. Other times, the literature provided additional perspectives on found themes. One main challenge for the researchers was to maintain a focus on customer needs and not be distracted by various operational aspects and features. Another reflection is that being open in observation is a balancing act when concurrently studying liter- ature about the phenomena under study. It is helpful to get help or inspiration from theory, yet one need to be open and not using theoretically preconceived categories. Being two researchers was a strength since it enabled an ability to continuously provide awareness of these issues. 3.4.6 Ethics Participant observation As previously mentioned the researchers were introduced to other members by the community managers when possible which provided access and legitimacy (Clancey, 2006). Another ethical consideration in the relationship to other members was that the researchers made sure to present themselves as researchers in line with Kawulich (2005). As suggested by Schensul and LeCompte (2012) all notes that included people were written using pseudonyms to ensure their anonymity and confidentiality. Interviews All interviewees were informed about the purpose of the study and how the results were being published. Bryman and Bell (2011) state that a credible rationale for participating in research should be provided to prospective informants. Furthermore the interviewees were assured confidentiality and informed about their right to exit at any time. Wallén (1996) stresses the importance of letting informants give their consent and freedom to exit a study if the wish to. All participants were asked for approval before recording the interviews. The recordings were shared between the researchers and stored safely; the files were kept out of online storage to minimise the risk of being attained by outsiders. When transcribing the interviews, names of 21 3. Methods persons and companies were anonymized to ensure confidentiality. 3.5 Quantitative methods 3.5.1 Kano-questionnaire The instrument used to measure identified needs was a self-completion questionnaire (Bryman & Bell, 2011). A Kano-questionnaire which can be found in Appendix C was designed from the voice of the customer and the participant observations. The paper by Berger et al. (1993) was used as guidance in designing the questionnaire, while affordance theory and the lens provided by Karahanna et al. (2018) was used to formulate the questions. More specifically the questions were phrased as affor- dances that could cater to the previously identified needs. A number of iterations of the questionnaire were designed and tested within the research team and oth- ers until a satisfactory design was reached. After that a pilot-questionnaire was sent out to the interviewees, asking for their feedback. As a result of the Covid-19 pandemic no on-site pilot-testing was done as was initially planned, yet some re- spondents were able to provide some feedback in person. The feedback was used to rephrase unclear questions, clarify the introduction and the answers. Several strate- gies were also employed to decrease the risk of a low response rate, such as making the survey relatively short, providing a clear rationale for participating, guarantee- ing confidentiality, clear instructions and attractive layout (Bryman & Bell, 2011). The respondents at A Working Lab were also provided with an incentive in form of a free lunch in the building’s restaurant. The survey was sent out to all members at United Spaces and A Working Lab in the form of a web link via email. The respondents had the option to choose whether to answer the survey in English or Swedish. The measured needs and their corresponding questions form the Kano-questionnaire can be found in Appendix C. Several trade-offs were present in the creation of the survey. The most apparent one relates to number and detail of questions versus respondent fatigue on the part of respondents. This fatigue refers to when a respondent becomes tired of answer- ing questions and therefore terminate the survey (Bryman & Bell, 2011). The aim becomes to capture as much as possible with a sensible number of questions and still maintain high validity. Therefore some aspects described clearly and coherently among interviewees and in literature was not included in the survey. This issue is present in the number of demographic questions asked as well. Since one aim was to detect possible segments asking the right demographic questions was of importance. Yet it is impossible to know beforehand what types of categorization is relevant for the given population resulting in a guessing game. Other studies and our findings were used as inspiration and material for discussion in the research group. The re- searchers believed that type of employment and one’s primary view of the coworking space might be a good predictor after some analysis of interviews and observations. 22 3. Methods Another aspect turning out to be quite troublesome was the wording of the possible alternatives of answers. Earlier research (Berger et al., 1993) as well as own early testing revealed some confusion regarding the alternatives. Therefore meticulous efforts were made to make the survey easy to follow and answer (Bryman & Bell, 2011) and the chosen answers can be found in table 3.2. Table 3.2 Wording of answers Original form Form used 1 I like it that way. I would really like it. 2 It must be that way. I would like it and expect it. 3 I am neutral. I would be neutral. 4 I can live with it that way. I would dislike it, but could tolerate it. 5 I dislike it that way. I would really dislike it. 3.5.2 Response rate United Spaces 1287 members were reached by the email containing the survey. After two reminders, a total of 94 persons had clicked the link to the survey and 86 persons had began answering the survey. After data cleaning, 55 complete responses remained. 6.7% = 86 1287 (3.1) A Working Lab 23 members were reached by the email containing the survey. After two reminders 15 persons had started answering the survey. After data cleaning, 12 complete responses remained. 67.2% = 15 23 (3.2) 3.5.3 Data cleaning and data analysis The survey responses were exported to an excel file and then transformed to fit the statistical analysis software JMP. The data was cleaned from respondents whose answers were exactly the same for almost every question. Those that had a faster response time than deemed possible to give a fair answer and persons that did not finish the survey were also removed before the analysis. Contingency tables within JMP were used as a way to explore potential explanatory factors from the demographic data. Furthermore, JMP was used to do a hierarchical clustering on the Kano-category response was done using Ward’s method. Both Excel and JMP was used to plot diagrams and charts as ways to explore the data. The Kano results were displayed as tables with the percentage of each Kano category for each respective question, for United Spaces as a whole, and each respective customer segment. A Working Lab was treated as a separate customer segment. As suggested by Matzler et al. (1996), S/DS-diagrams were used as a way to understand the customer needs. 23 3. Methods 3.5.4 Validity While designing the questionnaire a lot of thought was spent on the issue of validity often provoked by the aspect of respondent fatigue, mentioned above. Questions were tested and reformulated to be able to capture the identified need or a part of it. The iterations of testing the survey provided feedback on clarity of questions and alternatives for answers as well. Yet individual respondents might interpret questions in new ways and answer differently than the same person would have if the question would have been understood as intended. There is also always an issue with surveys in that individuals might provide answers that contradicts how they would feel about something if experiencing the same thing in real life. Another issue relates to how the combinations of answers are supposed to be inter- preted as classifications according to the Kano evaluation table (table 3.3). Specif- ically two combinations have provoked our thoughts. The first one is where a re- spondent gives the first alternative as an answer to the functional question and the second alternative to the dysfunctional one (F1-D2). This is classified as an at- tractive quality, yet to answer must-be for a dysfunctional formed question is not entirely logical. This might be a minor issue, since this combination seldom occurs. Table 3.3 Original Kano evaluation table, reproduced from Berger et al. (1993) Customer Requirements Dysfunctional 1 2 3 4 5 like must-be neutral live with dislike Functional 1 like Q A A A O 2 must-be R I I I M 3 neutral R I I I M 4 live with R I I I M 5 dislike R R R R Q A more problematic issue is the combination F2-D4. Building on the theory behind the questionnaire this could also be interpreted as a one-dimensional quality. The respondent answers that it is a must-be that a certain aspect is provided and he or she can live with a situation where it is not provided, yet this is not optimal. This is consistent with a satisfaction proportional to the level of provision, yet not as clearly as in the combination F1-D5. In our pilot study we noticed that the combination F2-D4 was quite popular yet the results showed no needs classified as one-dimensional. We saw this as a sign that the combination F2-D4 should be evaluated as one-dimensional and changed the original one (table 3.3) to the one that can be seen in table 3.4 24 3. Methods Table 3.4 Developed Kano evaluation table, developed from Berger et al. (1993) Customer Requirements Dysfunctional 1 2 3 4 5 like must-be neutral live with dislike Functional 1 like Q A A A O 2 must-be R I I O M 3 neutral R I I I M 4 live with R I I I M 5 dislike R R R R Q 3.5.5 Ethics Kano-questionnaire When it came to the questionnaire the respondents were informed that the data would be handled in accordance with GDPR, that their responses would be anony- mous, and that they had to give their consent before being able to proceed. Fur- thermore the respondents were informed that the research team were the only ones who would have access to the data. In the end of the survey the respondents were given the option to provide their email address if they wanted to have a copy of the final report. Before providing their email address they were informed that by doing so, the survey would no longer be anonymous. 25 3. Methods 26 4 Literature review Scholars have now for almost a decade tried to understand the coworking phe- nomenon through various perspectives. Albeit many notable contributions, the body of literature surrounding the phenomenon is not a particularly distinct one. With this note, we here outline some main themes found in the literature and answer the first research question. 4.1 What is coworking? Aside the attempts to find short and descriptive definitions of coworking, the phe- nomenon is portrayed in terms of features or service attributes, forms and typologies as well as proposed benefits. Some authors also develop novel theories in pursuits to capture new perspectives. 4.1.1 Definitions The one thing that clearly unites scholarly definitions of coworking is the aspect of co-location, the sharing of space (Bilandzic & Foth, 2013; Capdevila, 2013; Gan- dini, 2015; Parrino, 2015). The spaces are characterized as localized (Capdevila, 2013), open-plan office environments (Spinuzzi, 2012) and its users are described as unaffiliated (Spinuzzi, 2012), independent (Capdevila, 2013), knowledge profession- als (Gandini, 2015). Definitions of coworking incorporates various ideas about the phenomenon as well: social learning, peer collaboration and creativity (Bilandzic & Foth, 2013), knowledge sharing (Capdevila, 2013) and even an atmosphere or a lifestyle (Moriset, 2013). Scholars note that defining the phenomenon is not an easy task. Spinuzzi (2012) made an early effort, by collecting and analyzing the definitions of proprietors and users, showing a great variety of definitions : ...beyond saying that they worked in the presence of other people, they provided definitions that were far from unanimous (p. 418). Parrino (2015) notices a continuum in where you may place coworking spaces in terms of allowing for interaction and collaboration. This aspect is incorporated into a quite comprehensive definition of coworking: 1. the co-localisation of various coworkers within the same work envi- ronment; 27 4. Literature review 2. the presence of workers heterogeneous by occupation and/or sector in which they operate and/or organisational status and affiliation (free- lancers in the strict sense, microbusiness, employees or self-employed workers) 3. the presence (or not) of activities and tools designed to stimulate the emergence of relationships and collaboration among coworkers. (Parrino, 2015, p. 265) Later, newer perspectives are offered; Bouncken and Reuschl (2018) shed light on the aspect of autonomy offered by coworking spaces in terms of utilization possibilities of space and infrastructure as well as communication. 4.1.2 Typologies Apparent from the short journey into definitions above, the coworking phenomenon seem to encompass heterogeneous views. This point becomes more vivid when regarding the various classifications of coworking spaces the literature. Spinuzzi (2012) employs an activity theory perspective and finds two configurations of coworking spaces: the good-neighbors configuration and the good-partners configu- ration. The former comprises of actors working in parallel with a collaborative focus on how best support everyone’s individual work, whereas the latter is the habitat of actors cooperating as partners (Spinuzzi, 2012). Kojo and Nenonen (2016) iden- tify coworking spaces in Finland using business model and level of user access to identify six typologies. Bouncken, Laudien, Fredrich, and Görmar (2018) classifies coworking spaces into four types and finds among them tensions regarding value creation and value appropriation relating to coopetition. Ivaldi and Scaratti (2019) focusing on the concept of sharing find four types of coworking spaces with different activities and views on sharing: infrastructure coworking, relational coworking, net- work coworking and welfare coworking. Fiorentino (2019) develops three typologies based on coworking spaces role in the socio-economic ecosystem: social incubators, start-up incubators and real-estate incubators. Apart from pure typologies, scholars identify various aspects that could differentiate coworking spaces from each other. Clifton et al. (2019) quantitatively investigate inputs, outputs and outcomes of coworking spaces. Ross and Ressia (2015) find var- ious characteristics: length of membership, type of ownership, hybrid space or not, degree of collaboration and similarity of members. Bouncken and Reuschl (2018) develop six dimensions of coworking: coworking users, social intensity, institution of the coworking space provider, physical assets, availability and professional focus and competition. Vidaillet and Bousalham (2018) find three dimensions of coworking spaces based on economic forms, relations and practices: the involvement, relational and exchange dimensions. Others views the phenomenon through a more political lens. Gandini and Cossu (2019) divide the emergence of coworking into three historical phases; they start with an avant-garde phase leading into amainstream "neo-corporate" phase to finally address a new "resilient" phase. 28 4. Literature review 4.1.3 Features When regarding features it is hard to overlook the aspect of co-location; it is a core feature of coworking. It is also being connected to various proposed benefits such as easing the process of socialization (Bouncken & Aslam, 2019), reducing transaction costs for collaboration (Capdevila, 2014) and generating organizationality (Blagoev, Costas, & Kärreman, 2019). There are other aspects that distinguishes coworking from other forms of shared offices than just the sharing of space. For example a focus on community and knowledge sharing (Capdevila, 2014) or collaborative activities (Waters-Lynch, Potts, Butcher, Dodson, & Hurley, 2016). Cabral and Winden (2016) even argue that co-location does not automatically lead to interaction or innovation; instead it requires applying appropriate strategic tools to emerge. How, then, is the space that is being shared designed? There is plenty of case studies reporting of space characteristics of their studied coworking spaces. Space design relates to other aspects of coworking: it can influence how encounters happens (Jakonen, Kivinen, Salovaara, & Hirkman, 2017), it can affect people’s attitudes to- wards a place and its users (Orel & Alonso Almeida, 2019) and it can stimulate user’s creativity (Yang et al., 2019). Cheah and Ho (2019) found that space creativity was positively related to the business model innovation outcome of member firms. In their study, respondents answered to what extent their work space design encour- aged creative thinking, playfulness and generated ideas of higher quality (Cheah & Ho, 2019). Spinuzzi (2012) makes a distinction between inward and outward facing space designs: ...the space design was inward facing, focused on facilitating comfort and relationships within the coworking site. ...the site was designed to be outward facing, to facilitate professional contacts with outsiders, to impress rather than to comfort. (Spinuzzi, 2012, p. 422-423) Location is another feature of coworking examined in many studies (Fiorentino, 2019; Wang & Loo, 2017; Lilja, 2019). Fiorentino (2019) found concentrations of cowork- ing spaces in Rome around the main infrastructure such as metro lines. Location is determined by aspects such as ideological mission and origin but accessibility unites all types of coworking spaces (Fiorentino, 2019). Both hosts and users of cowork- ing regard location close to city center or metro stations as very important (Wang & Loo, 2017) and accessibility is even regarded as the most important reason for choosing a coworking space (Weijs-Perrée, van de Koevering, Appel-Meulenbroek, & Arentze, 2019; Capdevila, 2013). Last but not least we will shortly mention the aspect of provided services. Often mentioned in this instance is wifi connection (Bilandzic & Foth, 2013; Spinuzzi, 2012), printer (de Peuter, Cohen, & Saraco, 2017; Ivaldi & Scaratti, 2019) and various forms of events (Brown, 2017; Cabral & Winden, 2016; Capdevila, 2014; van Dijk, 2019). Mariotti, Pacchi, and Di Vita (2017) make a distinction between traditional services (administrative offices and meeting rooms) and digital services (wifi connection and printers). 29 4. Literature review 4.1.4 Proposed benefits The arguments for coworking are many, ranging from providing a sense of community and collaboration to increasing productivity and knowledge sharing. For a fuller picture, see table 4.1. These proposed benefits are in various ways interdependent, but we display them as more distinct entities. Through coworking, sociality can be provided to its’ users creating relationships leading to a concept of community (Parrino, 2015). Spinuzzi, Bodrožić, Scaratti, and Ivaldi (2019) develop two concepts of community within coworking, namely gesellschaft communities and collaborative communities. In the former, relationships between users are institutional or transactional, while they are networked relating to common project objectives in the latter (Spinuzzi et al., 2019). Garrett, Spreitzer, and Bacevice (2017) develop three types of collective actions that can build a sense of community in coworking spaces: endorsing, encountering, and engaging. Users collectively endorse a vision of community while leaving room for everyone to shape the vision to one’s individual needs (Garrett et al., 2017). Users can both passively encounter a sense of community or actively engage in co-constructing the sense of community (Garrett et al., 2017). Rus and Orel (2015) add to the discussion by claiming community building as a process in need of careful management and time. Collaboration is another aspect often connected to coworking (Butcher, 2018; de Peuter et al., 2017; Justin, 2019; Yang et al., 2019) and Salvador et al. (2018) view coworking spaces as excellent for collaboration. Capdevila (2014) develops three collaborative approaches: cost-based collaboration with the goal of reducing operational or transaction costs, resource-based collaboration where users aim to access new knowledge and resources and relational collaboration relating to a search for synergies and building the community. Castilho and Quandt (2017) find four dimensions available to founders and managers to build collaborative capability in coworking: knowledge sharing, enhancing a creative field, supporting individual actions for collective results and supporting collective action towards an effective execution. This seems to indicate what Ross and Ressia (2015) argue, namely that collaborative behaviour does not appear automatically by co-location and indeed Parrino (2015) stresses the importance of facilitators of collaboration and being part of networks outside coworking. Some benefits are more closely linked to entrepreneurs and freelancers such as sup- port, status and managing the boundary between work and leisure. De Peuter et al. (2017) claim "coworking spaces function as infrastructure for mutual aid" (p. 700) and relate it to having support similar to individuals in organizations. Support is related to giving feedback and exchanging help in struggles (de Peuter et al., 2017), events and activities offered by proprietors (Bouncken & Reuschl, 2018) and sharing knowledge and develop business skills (Butcher, 2018). Gerdenitsch, Scheel, Andor- fer, and Korunka (2016) stress the importance of social support and find a positive relation between this support and performance satisfaction. Coworking spaces pro- vide organizationality, meaning that they can help managing the boundary between work and leisure (Blagoev et al., 2019) and give structure to work and work-hours (Bouncken & Reuschl, 2018). Spinuzzi (2012) notices that many coworkers had 30 4. Literature review problems with distractions, self-motivation and isolation when working from home. Bouncken and Reuschl (2018) summarize some of these aspects: The social and material resources of coworking-spaces can support en- trepreneurs and micro-companies in their decisions. Coworking can... offer solutions for their problem of isolation, which impacts psyche, physique, and professional performance. (Bouncken & Reuschl, 2018, p. 328-329) Van Dijk (2019) also brings in the aspect of coworking spaces’ ability to provide status for its users. Coworking is also argued to provide co-discipline (Blagoev et al., 2019), increase performance (Bouncken & Reuschl, 2018) and productivity (Blagoev et al., 2019; Clifton et al., 2019; Salvador et al., 2018; Yang et al., 2019). Bouncken and Reuschl (2018) specifically find that performance "improves by the learning processes among coworking-users that take upon the individual efcacy, trust and community among coworking-users." (p.317). Salvador et al. (2018) find that improvement of produc- tivity is related to social interactions and the coworking spaces environment. Innovation is another proposed benefit of coworking (Yang et al., 2019) and Clifton et al. (2019) find enhanced level of innovation reported from users. Cheah and Ho (2019) see a positive relation between space creativity and business model inno- vation outcome. Cabral and Winden (2016) describe how four strategies can be employed by coworking spaces to encourage interaction and innovation: coworking space management as a connector, regulating the mix of workers, interior design for interaction and tools for networking. Bouncken et al. (2018) argue that coworking spaces can utilize a positive tension between collaboration and competition to build the core of innovation ecosystems. As noticed above, social interactions are mentioned and connected to other proposed benefits, for example to productivity (Salvador et al., 2018) and innovation (Cabral & Winden, 2016). Bouncken and Reuschl (2018) see increasing social interaction as a central function to the phenomenon and relate it to community. Jakonen et al. (2017) even argue that encounters, intended and unintended, are a form of production. However encounters do not necessarily take place in spite of intentions to increase serendipitous encounters (Jakonen et al., 2017). Other benefits mentioned are creativity (van Dijk, 2019; Yang et al., 2019), knowl- edge sharing (Bouncken & Aslam, 2019; Salvador et al., 2018; Parrino, 2015) and social well-being (Yang et al., 2019). 4.2 Who is the user? Coworking is early described as an alternative to home-based work (Ross & Ressia, 2015). It is therefore often related to freelancers, entrepreneurs, knowledge workers, nomadic workers and self-employed workers (Waters-Lynch & Potts, 2017; Vidaillet & Bousalham, 2018; van Dijk, 2019; Merkel, 2019). Early statistics confirms this 31 4. Literature review Table 4.1 Proposed benefits of coworking Collab orat ion Community Crea tiv ity Innovat ion Knowled ge sh ari ng Man age boundary betw een work & lei su re Producti vit y & perf orm an ce So cia l in ter act ion So cia l w ell- bein g Sta tus Su pport ( so cia l & perf orm an ce) Blagoev et al. (2019) x x x Bouncken & Reuschl (2018) x x x x Bouncken et al. (2018) x x Bouncken & Aslam (2019) x Brown (2017) x x Butcher (2018) x x x Cabral & Winden (2016) x x Capdevila (2014) x Castilho et al. (2017) x Cheah & Ho (2019) x Clifton et al. (2019) x x de Peuter et al. (2017) x x Garret et al. (2017) x Gerdenitsch et al. (2016) x Merkel (2019) x Orel et al. (2019) x Parrino (2015) x x Ross & Ressia (2015) x Rus et al. (2014) x Salvador et al. (2018) x x x x Spinuzzi (2012) x x x x x Spinuzzi et al. (2019) x van Dijk (2019) x x Walden (2019) x Yang et al. (2019) x x x x x x 32 4. Literature review with 55 percent of coworkers being freelancers in 2012, a share that shrinks to 42 percent in 2019 (Deskmag, 2019). Clifton et al. (2019) mention that employees of larger organizations might increasingly be motivated to cowork, and uses the term "location-independent work" to describe the potential users of coworking. Parrino (2015, p.265) organizes coworkers into three groups: 1. freelancers in the strict sense; 2. microbusinesses that are based in the coworking spaces; 3. employees or self-employed workers, whose activity is done on behalf of a company based outside the coworking spaces. Parrino (2015) speculates that the reason for the third group could be related to strategic and management needs of organizations or the needs of their workers. Brown (2017) distinguishes and gives detailed descriptions of 4 groups of coworkers: mentees, networkers, motivators and reluctant soloists. 4.3 Why cowork? Numerous efforts have been made to capture the reasons why people use coworking spaces and while the results vary in breadth and depth they hopefully paint a quite comprehensive picture together. Interaction and social support are mentioned by many (Butcher, 2018; Clifton et al., 2019; Ross & Ressia, 2015; Spinuzzi, 2012; Weijs-Perrée et al., 2019). Brown (2017) sees strong support for a need to interact with "like-minded" professional and Jakonen et al. (2017) found the main motivation for joining was to be part of a social context. A closely related wish is to be part of a community (Garrett et al., 2017; Weijs-Perrée et al., 2019). Another common theme is to escape isolation in one’s current work situation (Clifton et al., 2019; Gandini, 2015; Garrett et al., 2017; Grazian, 2019; Jakonen et al., 2017) or as Brown (2017) describes it: "...an antidote to professional isolation" (p.120). There is a number of studies finding an expressed need for a more professional work environment (Brown, 2017; Robelski, Keller, Harth, & Mache, 2019; Weijs-Perrée et al., 2019). This can be in form of hopes of acquiring a professional status associated with a certain space (van Dijk, 2019; Gandini, 2015) and to impress outsiders (Spinuzzi, 2012). Many of the aforementioned aspects can also be related to a need for a separation of home and work life (Brown, 2017; Ross & Ressia, 2015; Weijs-Perrée et al., 2019; Clifton et al., 2019). Among other more frequently mentioned motivations are collaboration (Ross & Ressia, 2015; Rus & Orel, 2015; Spinuzzi, 2012), access to a network (Clifton et al., 2019; van Dijk, 2019; Gandini, 2015; Weijs-Perrée et al., 2019), professional support (Brown, 2017; Butcher, 2018; Spinuzzi, 2012; Weijs-Perrée et al., 2019) affordable accommodation (Butcher, 2018; Clifton et al., 2019; Ross & Ressia, 2015; Weijs-Perrée et al., 2019) and flexibility (Ross & Ressia, 2015; Spinuzzi, 2012; Weijs- Perrée et al., 2019). These aspects and some additional ones can be viewed more clearly in table 4.2. Most studies are qualitative and few make strong statements regarding relative im- portance of motivations, yet Clifton et al. (2019) measure this aspect quantitatively 33 4. Literature review in a questionnaire with 76 respondents at two coworking spaces in Wales, UK. Overcoming isolation is the most common of negative aspects (40%), while posi- tive aspects receive higher agreement: creative environment (45%), cheap business solution (45%), network (50%) and interaction (50%). Weijs-Perrée et al. (2019) (see figure 4.1) receive 219 responses to a survey sent out to 25 coworking spaces in the Netherlands where looking for a place outside home (65%) and vibrant and creative atmosphere (55%) where the most important motivations with affordable accommodation (37%) and social interaction (32%) at third and fourth place. Motivat ions to work at a cowork in g space (N=219) The possibility for work related conversations with other co-workers I was looking for a workplace outside my home Vibrant and creative atmosphere in the co-working Affordable accommodation Social interaction with co-workers The opportunity to network with co-workers Feeling of being part of community Flexibility Professional supportive work environment Professional appearance for the company 1st motivation 2nd motivation 3rd motivation 0 20 40 60 80 100 120 140 160 Survey done in the Netherlands Number of respondents Figure 4.1 Motivations to work at a coworking space, reproduced from Weijs-perrée et al. 2019 34 4. Literature review Table 4.2 Motivations for coworking Brown (2017) Butcher (2018) Clifton et al. (2019) Gandini (2015) Garret et al. (2017) Grazian (2019) Jakonen et al. (2017) Robelski et al. (2019) Ross & Ressia (2015) Rus et al. (2014) Spinuzzi (2012) van Dijk (2019) Weijs-Perrée et al. (2019) Access to network x x x x x Affordable accommodation x x x x Avoid isolation x x x x x x Being part of community x x Collaboration x x x x Flexibility x x x Inspiring and creative atmosphere x x x Interaction and social support x x x x x x x Location x Nature of precarious work x Productive work environment x x x x x Professional support x x x x Professional work environment x x x x x x Sense of ownership x Separate home form work x x x x Sharing ideas and knowledge x x 35 4. Literature review 4.4 What aspects are important? Clifton et al. (2019) (see figure 4.2), when asking respondents why they chose their specific coworking space, report around 50 percent of coworkers selecting commu- nity and interaction, while the most common responses were social and enjoyable atmosphere and good office infrastructure (both around 60%). Weijs-Perrée et al. (2019) report accessibility by car and public transport being the most important attribute followed by atmosphere and interior aesthetics. Older coworkers (>35 years) and managers rates a higher importance in accessibility by car while younger seem to have a higher tendency towards public transportation (Weijs-Perrée et al., 2019). Another notable aspect in the results are that coworkers on average pre- fer a homelike environment over a more modern one, but individuals with higher education shows higher preference for modern interior (Weijs-Perrée et al., 2019). Seo, Lysiankova, Ock, and Chun (2017) received 56 responses from coworkers in South Korea which showed that users thought relationship facilitation was the most important aspect, meaning being able to naturally encounter other coworkers when one wants to. Other important aspects were service diversity, price plan, networking event and party in descending order (Seo et al., 2017). Why did you choose your current cowork ing space? (N=76) Possibility to work in groups Random discoveries and opportunities Flexible work times Knowledge sharing A close distance to my home Good value for money It is a community Interaction with others A social or enjoyable atmosphere Good office infrastructure 0 10 20 30 40 50 60 70 (% of respondets selecting) Survey done in Wales, UK Figure 4.2 Why did you choose your current coworking space? reproduced from Clifton et al. 2019 4.5 Negative aspects The literature portrays a quite positive picture of coworking, yet some negative aspects are mentioned. Ambivalence from conflicts arising from shared work, specifically regarding capture and commodification of shared resources and fair acknowledgement of work (Waters- Lynch & Duff, 2019). Tensions relating to separating contributions in shared value 36 4. Literature review creation, as well as tactics concerning ideas and knowledge are related to coopetition (Bouncken & Aslam, 2019). Unintended leakage of ideas and knowledge is argued to impair learning and entrepreneurial performance, causing reduction of trust and community building (Bouncken & Reuschl, 2018). Bilandzic and Foth (2013) finds barriers for social learning since users find it hard to identify or approach other unknown users and remain unaware of each other’s expertise. Jakonen et al. (2017) also note that users avoid or neglects encounters due to a primary focus on their own work. Another issue for coworkers is to not be distracted from work activities due to time and energy spent on networking (Bouncken & Reuschl, 2018). Capdevila (2013) mentions a difficulty for coworkers to separate their private and professional spheres, noting that coworkers often act concurrently as individuals and the firms they represent. Spinuzzi (2012) argues that tensions can emerge in cowork- ing spaces since contradicting configurations (described above) are superimposed in the same space. 37 4. Literature review 38 5 Qualitative results & discussion In this section we outline the customer needs found in the collected data and ground them in the SDT framework and previous research. It is Worth noting that cowork- ing spaces are shared not only by freelancers and entrepreneurs. Members are em- ployed by companies of various sizes, ranging from less than ten employees to more than 250. 5.1 Framework When relating SDT in the coworking setting it became evident that the needs could not only be explained in relation to the self. Therefore, four different levels were added to the framework as a way to discuss and relate the identified needs. Self Needs relating to the self are those that are found in SDT literature. It is about the individual’s personal needs. Peers The needs relating the peers of the coworking space are those that exist in the interface between members. It is about business relations, but also between members as individuals. Host Some of the identified needs are in relation between the hosts and members. Space The fourth level is in relation to the coworking space, the spatial design, number of rooms, and its location for example. 39 5. Q ualitative results & discussion a31 a32 Self Peers A6 A4 A9A7 A2 A8 Relatedness Autonomy Competence R3 R4 R1 A1 C3 C8C4 C5 C1 Host R2 A5 C9 C7 C2 Space C10 C6A3 Attractive One-dimensional Must-be Indifferent Kano classification 1 12 # of interviewees that mentioned need (out of 13) r11 r12 R5 Figure 5.1 Uncovered customer needs in framework 40 5. Qualitative results & discussion Table 5.1 Label and the corresponding need Label Need A1 To know which other companies are members and what they do A2 To be able to market one’s business A3 To be able to choose a suitable work area a31 Access to whiteboard a32 Rent conference/meeting rooms A4 To be transparent when meeting others A5 To be able to scale one’s business A6 To have phone calls or conversations, without disturbing others A7 To be able to manage confidential information safely A8 To feel in control of social interactions A9 To be able to focus on work activities C1 To learn new things from peers and events C2 To be healthy C3 To meet people that can lead to business opportunities C4 To cooperate/collaborate with relevant actors C5 To be able to receive help or input from others C6 To have a workplace that gives you energy C7 To be able to work smoothly without technical disruptions C8 To be able to share knowledge/ competence C9 To be able to focus on the core business C10 To feel an increased productivity from one’s workplace R1 To belong to a community r11 Meet new friends r12 Have small chats with members R2 To feel like a prioritized customer R3 To feel welcomed at one’s workplace R4 That one’s workplace leaves a good impression on guests R5 To have a workplace at a fixed location 41 5. Qualitative results & discussion 5.2 Relatedness Relatedness is about the reciprocal caring and concern between individuals and groups of people (Ryan & Deci, 2017); "To love and care, and to be loved and cared for" (E. L. Deci & Ryan, 2000, p. 231). 5.2.1 To belong to a community (R1) To belong to a community has emerged as a need among interviewees. Some express a wish to be part of a community, yet remain passive. Others take it upon themselves to make community belonging emerge. Immersion and interviews indicate that a sense of community will not emerge from co-location and sharing of space alone. It appears that it has to be facilitated. On some mornings I tell those that I bump into that I would like to get to know new people, and that those who feel the same are welcome to join me at a table later (Self-employed) Some people do not have the need to be a part of a community at their coworking space but appreciate if the possibility exist, while some seem to be indifferent all together. Within a sense of community there is an inherent need to relate to other people, but the expression of this need varies between individuals. Some interviewees see social relationships as a goal in itself and tries to satisfy the need by having social interactions with other members. I could have chosen to work from home, but I felt that I wanted to be a part of a community. Otherwise you’ll easily become quite lonely when self-employed... (Self-employed) Some members get their social interaction from hosts. This was confirmed from observations as well. Members would often stop and chat with a host while grabbing a coffee and multiple members expressed enjoying small chats with people in the reception. Sometimes, instead of using the app to book a meeting room, I go to the reception and ask them to help me book a room, because it’s nice to have a small chat. (Employed by organization) Others get their social interactions from their colleagues at the coworking space, and some meet friends during lunch. There are members that want deeper social relationships with their coworking peers. As one member puts it, being asked what relationship that member hoped to achieve: Friendship - that’s the maximum you can achieve. Exchange numbers and hang after work. (Employed by organization) 42 5. Qualitative results & discussion To relate to others has a strong connection to the relatedness dimension of SDT. It is also a common theme in previous literature where several authors put it in the category of being part of a community. For example, Jakonen et al. (2017) found that the main motivation for joining a coworking space was to be a part of a social context. As noted in previous research the need for social support and interaction is one of the main reasons for joining a coworking space (Spinuzzi, 2012; Brown, 2017). Our findings illuminated that the need to relate to others at work varied between individuals. Consequently they might prefer different affordances that satisfies this need. Some wanted deep and meaningful relationships such as friendship, while others felt content with a small chat with the hosts. In line with the findings of Cabral and Winden (2016), a sense of community will not emerge by co-location alone; it has to be facilitated by other methods. Our study and previous studies note that such interactions might not take place in spite of intentions to do just that (Jakonen et al., 2017). Members employed by companies with multiple members at a coworking spaces usually got their social interaction from their colleagues. Self-employed entrepreneurs, on the other hand, had to get their interactions from other members or hosts. At first glance, the need to relate to others seems straightforward. But the different attitudes towards the need and ways of satisfying it, might create tensions and dynamics. Those who crave deeper relationships could end up disappointed by the lack of need-fulfillment, while those who only want to have a small chat might end up annoyed when feeling pressured to interact. 5.2.2 To feel like a prioritized customer (R2) Several members expressed a need to feel like a prioritized customer. Some phrased it as the hosts attention to detail, by doing the right thing at the ri