New Service Development Process:

(1)

0

New Service Development Process:

A Multiple-case study on service innovation

projects in the hospitality industry

By

Manon Eisses

Master Thesis

MSc. BA Strategic Innovation Management

University of Groningen

Faculty of Economics and Business

Date: 20

th

of June, 2017

Supervisor: dr. J.D. van der Bij

Co-assessor: dr. W.G. Biemans

Manon Eisses

m.v.eisses@student.rug.nl

Student number: s3033996

(2)

1

ABSTRACT

The service innovation literature demonstrates that most researchers applied the knowledge of NPD to the domain of NSD, and so far a generally accepted NSD model is still missing. Therefore service innovation literature is in need of an alternative description of the NSD process, which is provided by this research from a pure service perspective. Through a multiple-case study research and cross-case analysis the study found the activities performed in the NSD process and persons that are involved. The research proposes an NSD process model consisting of 8 phases which includes 16 aggregated activities in total.

(3)

2

INTRODUCTION

Services, as argued by several researchers, tend to be intangible, heterogeneous, simultaneously produced and consumed, and perishable (e.g. Berry, 1980; Zeithaml, 1981; De Brentani, 1991), and can be described as “a deed, process, and performance” (Zeithaml & Bitner, 1996, p. 5). Currently, service sectors form a considerable part of the world economy. In 2015, 62.6% of the world‟s gross domestic product (GDP) was represented by service sectors (CIA WorldFactbook, 2015). The world‟s most advanced economies are especially dominated by services, as it often generates more than 70% of their GDP (Cusumano, 2010; Ostrom et al., 2010); for example, in 2015 The Netherlands generated 70.2% through services, France 78.8%, the European Union as a whole 70.6%, and the United States 79.5% (CIA WorldFactbook, 2015). Due to the increasing importance of services and the awareness that innovation is important for success, survival and renewal of organizations that are located in highly competitive markets (Brown & Eisenhardt, 1995), and hence, that new services may bring forward a competitive advantage, the literature on service innovation is growing (Biemans, Griffin, Moenaert, 2016). Despite the growing literature on service innovation, it is seen that in contrast to the logical assumption that most of all innovation research would be on service innovation, the vast majority of innovation studies focus on products instead of services (Storey, Cankurtaran, Papastathopoulou, & Hultink, 2016).

In associated literature a service innovation has been defined in several ways. This paper goes with the more common tradition and follows Biemans, Griffin, and Moenaert (2016) who consider „service innovation‟ and „new service development‟ (NSD) as synonyms, which is defined as “the process of

devising a new or improved service, from idea or concept generation to market launch” (Biemans et

al., 2016, p. 383). Avlonitis, Papastathopoulou and Gounaris (2001) proposed that the performance outcome of a new service is influenced by the followed development process of that service. The NSD process exists of three components: „what‟, „how‟, and „who‟ (Avlonitis et al, 2001). The „what‟ component contains the NSD activities that lead to the creation of new services, the „how‟ component considers how the process is performed in terms of three process formality dimensions - systematic behavior, documentation, and assignment of responsibilities - and the „who‟ component is about cross-functional involvement (Avlonitis et al, 2001).

(5)

4 other NSD topics. Also Biemans et al. (2016) found that the topic NSD process and execution declined over time, and that other topics such as „strategy‟ and „co-development and alliances‟ gained more interest in the closing era of research that they investigated (2009-2012).

Thus, most research that has been performed on the process of NSD is not very recent. Moreover, several researchers have pointed at the need for further research on NSD as it remains underdeveloped and not well understood (e.g. De Jong & Vermeulen, 2003; Droege et al., 2009; Storey & Hull, 2010). De Jong and Vermeulen (2003, p. 854) state that “future research should focus more often on

alternative descriptions of the NSD process” as there exists hardly any evidence that service

innovations go through certain specific stages. It was found by Biemans et al. (2016) that most of the researchers applied the concepts, frameworks, and methods that are used to understand NPD to the domain of NSD (e.g. Bowers, 1989; De Brentani, 1991). However, according to Storey et al. (2016, p. 542) “it would be wrong to treat the development of new services and new products as the same”. There is a lack of coherence in knowledge on the topic of the NSD process, and Biemans et al. (2016, p. 395) argue that “managers who want to start or improve their service offerings will find only limited

help from the available NSD literature”. Therefore, they state that an accepted model of the NSD

process must be developed (Biemans et al., 2016). This research paper will focus on the service development process from a service perspective and will include the „what‟, „how‟, and „who‟ components as described by Avlonitis (2001), in order to come up with an alternative description of the NSD process. Therefore the following research question is established:

How does the NSD process look like from idea or concept generation to market launch?

In order to fully answer the research question, the following sub-questions are set up: 1) What are the

activities in the NSD process that lead to the creation of the new service, and in what order are they being performed?; 2) How is the process performed in terms of systematic behavior, documentation, and assignment of responsibilities?; 3) Who are involved in the NSD process?

(6)

5 effect of the life-cycle and teleological perspective (Van de Ven & Poole, 1995) on the NSD process and concludes that the NSD process in general moves more towards a teleological perspective.

The outcome of this research is relevant for managers of service organizations, especially those in hotels, who are developing service innovations. In addition, the service innovation literature and the NSD domain in specific, will benefit from this research as it enhances the current understanding by researching the innovations from a service perspective rather than having the NPD process in mind. It gives insight into how businesses, hotels in this case, develop new services. Moreover, the NSD domain can benefit from this in-depth case study as a shift in research approaches to more exploratory research is seen as beneficial by Biemans et al. (2016). Lastly, this study contributes to the service innovation literature as explaining NSD processes by applying the teleological- and life-cycle perspective is only recently introduced by Biemans et al. (2017) and so far rather limited.

After the introduction of this research paper‟s topic, the concepts are further defined in the theoretical section. Subsequently, in the next chapter, the methodology that is used for this study is explained. Thereafter the findings of the case studies are described and discussed in detail, where a new model for the NSD process is proposed. The paper closes by discussing the theoretical contributions, managerial implications, and limitations and giving recommendations for future research.

LITERATURE REVIEW

This chapter demonstrates what has already been written in the literature about the NSD process. The first part is structured according to the „what‟, „how‟, and „who‟ components, as described by Avlonitis et al. (2001), that the company manages in the NSD process. Furthermore, insight is given in what has been written on service innovation in the hospitality industry. Lastly, the life-cycle and teleological perspective on NSD are discussed, as they clarify the progress of the NSD process.

The New Service Development process

‘What’ - NSD process stages and activities

(7)

6 recognized the uniqueness of services and developed an extended model of fifteen stages, see table 1. Avlonitis et al. (2001) observed several models for new products and new services, and said that most cases involve the following 5 phases: idea generation and screening, business analysis and marketing strategy, technical development, testing, and commercialization/launching. They established a list of 29 specific development activities, which were subdivided under the above-mentioned five phases. The model of Scheuing and Johnson (1989) includes more testing activities, a personnel training and a post launch review compared to Bowers (1987, 1989). A recent study of Biemans et al. (2017) confirms the importance of testing with their finding that idea generation, screening, and testing are important phases for both incremental and radical NSD. They argued that these phases are crucial to all new service developments, independent of the service‟s degree of intangibility or newness to the market. Avlonitis et al.‟ (2001) list of activities also shows several tests to be performed. Moreover, it consists of receiving feedback after the service launch and taking corrective actions, which are very similar to a post-launch review as introduced by Scheuing and Johnson (1989). Moreover the model of Alam and Perry (2002) also includes a testing phase, and just as Scheuing & Johnson‟s model (1989) a specific phase for personnel training (table 1).

Bowers (1987, 1989) Scheuing & Johnson (1989)

Avlonitis (2001) Alam & Perry (2002) 1. Develop a business strategy 2. Develop a new services strategy 3. Idea generation 4. Concept development and evaluation 5. Business analysis 6. Service development and evaluation 7. Market testing 8. Commercialization 1. Formulation of new service objectives 2. Idea generation 3. Idea screening 4. Concept development 5. Concept testing 6. Business analysis 7. Project authorization 8. Service design and

testing

9. Process and system design and testing 10. Market program

design and testing 11. Personnel training 12. Service testing and

pilot run 13. Test marketing 14. Full scale launch 15. Post launch review

1. Idea generation and screening

2. Business analysis and marketing strategy 3. Technical development 4. Testing 5. Launching 1. Strategic planning 2. Idea generation 3. Idea screening 4. Business analysis 5. Formation of cross-functional team 6. Service design 7. Personnel training 8. Service testing and

pilot run 9. Test marketing 10. Commercialization

Table 1 – Existing NSD models in the literature

(8)

7 other facilitates the ability to perform some stages in parallel in order to accelerate the process (Alam, 2007), with which they were the first explaining which exact stages could be combined. The phases that can be performed simultaneously are strategic planning and idea generation, idea screening and business analysis, and lastly, personnel training and service testing/pilot run (Alam & Perry, 2002). Both models consist of ten identical development stages, see table 1. Compared to the other models it is the only one with a specific stage for forming cross-functional teams (Alam, 2007). Alam and Perry‟s (2002) model is rather similar to the stage-gate model for NPD of Cooper (2001); the innovation process is also divided in stages which exist of multiple, often parallel, activities, and it also has gates which are intended as quality control checkpoints for which certain criteria should be met in order to proceed the innovation project to the next stage (Alam, 2007).

Furthermore, several researchers mentioned that involving customers in the service development process may be more helpful than when developing products (e.g. Martin & Horne, 1995; Alam & Perry 2002; Edvardsson et al., 2006). Opportunities for service innovations are often discovered during the interaction with customers (Kindström, Kowalkowski & Sandberg, 2013), and collecting data about them is especially important as a basis for new service development (Edvardsson et al., 2006; Kindström et al., 2013). Besides, customer involvement positively and directly affects the technical quality of the innovation as well as the innovation development speed (Carbonell, Rodríguez-Escudero, & Pujari, 2009). Carbonell et al. (2009) concluded that customer involvement is important in technological uncertain conditions and that customers should be involved throughout the whole service development process, because the impact on the new service‟s performance demonstrated to be independent of the stages in which customers were involved. But, the customer‟s role should be well understood, as the customer fulfills both the role of customer as well as that of a co-producer in services (Martin & Horne, 1995). So far only Alam and Perry (2002) proposed a model in which customer input is embodied at the different stages. For example, in the idea generation phase customers could state their needs and identify problems for which solutions can be proposed. Moreover, in several phases they could participate and give feedback (Alam & Perry, 2002).

‘How’ - NSD formality dimensions

(9)

8 development process (Avlonitis et al., 2001). Project teams can divide the various tasks in order to increase efficiency and effectiveness (De Jong & Vermeulen, 2003).

Johne (1984) and Johne and Harborne (1985) concluded that during the initiation phase of the NSD process both active- and less active innovating companies have a loose formality structure, however when the project approaches the implementation stage, companies that actively innovate show a tighter formality structure, while companies that innovate less actively mostly preserve a loose formality structure. Avlonitis et al. (2001) found in their research that the degree of formality in the structure depends on the type of new service and the degree of innovativeness. Hence, the development process has not an equal formality structure for all new service innovations.

‘Who’ - NSD cross-functional involvement and the role of key persons

The third block of NSD variables that Avlonitis et al. (2001) refers to as the „who‟ component is about cross-functional involvement. The creation of cross-functional teams directly contributes to the new services development‟s effectiveness. Its success in service development activities comes from the ability to combine knowledge and increase problem-solving capacity when impediments arise in the process (Gallouj & Weinstein, 1997). On the other hand, processes without cross-functional teams may suffer from departmentalized structures that hinder NSD (De Jong & Vermeulen, 2003). Avlonitis et al.‟s (2001) study showed that the degree of cross-functional involvement differs across service innovation types, however no conclusion about this was made in their article.

(10)

9

Service innovation in the Hospitality Industry

It is recognized that service innovation capability is also important for hotels to respond to customer expectations and create unique service innovations, in order to gain a competitive advantage (Tang, Wang, & Tang, 2015). Kandampully, Bilgihan and Zhang (2016) concluded in their research that in competitive markets hospitality firms are not different from other industries: “They must adopt

strategies to support creative innovation related to the product, services, and experiences offered to customers” (Kandampully et al., 2016, p. 160). In their research they highlighted the critical role of

people inside hospitality firms and mentioned the importance of employee-centricity. Creativity of people has become a strategic priority of many leading hospitality firms, as technology on its own is not enough for innovation (Kandampully et al., 2016). Bowen (2016) argued that hospitality firms should invest in recruiting employees that have the ability to read and understand customers, so that their needs could be revealed. Besides, employees should be motivated by creating work conditions that encourage and reward them for generating creative ideas that could improve the hospitality firms‟ service (Bowen, 2016). This would result in a people-technology hybrid organization that allows the hospitality firms to be truly creative and innovative, as people‟s creativity and knowledge from using technology, by and from both employees and customers, is used (Kandampully, et al., 2016).

Life-cycle and teleological perspective on NSD

(11)

10

METHODOLOGY

This study is based on the explanatory paradigm as the purpose of this study was to produce descriptive knowledge on how the NSD process is performed and explanatory knowledge on which factors in the NSD process are important to produce successful service innovations (Van Aken, Berends, & Van der Bij, 2012). A theory development approach is most appropriate for this study, as a general problem exists that is familiar to many companies on aggregated level (Van Aken et al., 2012); namely that organizations find limited guidance in the literature on NSD when developing new services (Biemans et al., 2016). The literature review shows that several authors discussed the NSD process stages, however so far there is not a widely accepted model for the NSD process yet. The literature that is purely based on NSD is still very exploratory in its nature, hence a theory development approach is suitable (Van Aken et al., 2012; Yin, 2014). By exploring this phenomenon in a pure service environment, a contribution could be made to the service innovation literature by providing an alternative description of the NSD process, as recommended by De Jong and Vermeulen (2003). In order to develop said new theory, a multiple-case study was performed to gather empirical evidence on the phenomenon, so that a cross-case analysis could be executed (Van Aken et al., 2012). This chapter describes the research design that has been selected, how data was collected, how the data was analyzed, and what measures were taken to ensure the quality of this study.

Research Design

To research the phenomenon from a different perspective and to provide a description of the NSD process, theory development was based on a multiple-case study design. A case study is a good method to accomplish the goal of providing a description (Eisenhardt, 1989). Particularly in this field of interest case study research is suitable, in order to gain understanding of how firms develop new services (Biemans et al., 2016). This research followed the process of building theory from case study research as described by Eisenhardt (1989) and is explained hereafter.

Case Selection

Since several students perform this research, the industry sectors were divided on the basis of everyone‟s preference. As previously mentioned, this study focuses on the hospitality industry, which includes hotels, contract food service, restaurants, pubs, bars, and night-clubs (Li et al., 2013). The author has chosen to focus on hotels. Four- and five-star hotels in the Netherlands that belong to a hotel chain form the population of this study. This industry is a perfect example of a service industry as it meets the characteristics of a service-based industry described by Cooper et al. (2008): intangibility, inseparability, and perishability (Li et al., 2013).

(12)

11

choose cases which are likely to replicate or extend the emergent theory” (Eisenhardt, 1989, p. 537).

This multiple-case study research examined the NSD setting at three different case companies; where in total five cases were investigated. This allowed replication logic as described by Yin (2014), in which cases are seen as independent experiments that confirm or disprove conceptual insights that have emerged. So, in this way a cross-case analysis could be executed (Van Aken et al., 2012). In order to be able to successfully compare the cases at a later stage, and to find out what activities and potential other elements matter to be able to successfully innovate services, the author made use of polar types at each case company: a successful service innovation project and an unsuccessful service innovation project. In this way the processes could be compared and differences could be revealed. Due to limited time of the author and as well as of the third case company, only one case (a successful service innovation) has been researched, which resulted in a total of five investigated cases. The detailed case selection criteria are displayed in table 2.

Case selection criteria

Hospitality organization type Hotels

Ownership/Management Belonging to a hotel chain

Ranking 4 to 5 stars

Location Netherlands

Study related requirements  (Service) innovation should be performed on a regular basis.  A successful service innovation and a service innovation

failure should be developed in order to compare cases.

 The projects should be performed in the recent past (in the last 0-5 years) and its success/failure should be confirmed.

Table 2 - Case Selection Criteria

(13)

12 that the projects that would best suit this research were chosen. Thereafter, follow-up interviews were scheduled with the project leaders; in two case companies these were the same persons, and in the third case company this was the Operations Specialist. During the data collection process also other persons that have been involved in a successful or unsuccessful project were contacted and interviewed. This was done to retrieve information from different angles as “variation is especially

important in theory building because it increases the broadness of concepts and scope of the theory”

(Corbin & Strauss, 2008, p.156).

Data Collection

In order to collect the data, data collection methods were combined; in-depth interviews were held for primary data and documents/archival sources were analyzed for secondary data. By using multiple data collection methods, a stronger substantiation could be provided and stronger propositions could be developed (Eisenhardt, 1989).

Interviews (primary data collection)

The primary data source was in-depth semi-structured interviews with people who have been involved in the selected service innovation projects. The interviews were held with persons having different responsibilities within the project. In this way more insights could be gathered as some persons only operated at the project level, whereas others were managing the project on the floor. Interviews were also held with employees who were actually present on the floor, perceiving the change and eventually providing the new service. In total 11 interviews were conducted, varying from two to five interviews per case company. Most interviews were held face-to-face on the location where the service innovation project took place and whenever this was not possible they were held via Skype or telephone. The average length of the interviews was 75 minutes, from which the shortest interview took 40 minutes and the longest interviews accounted for about 1 hour and 45 minutes.

(14)

13 summaries and transcripts case descriptions were written, which were also sent back to the main contact persons of the three case companies. Again, these persons could provide feedback and remarks, and the construct validity of this research was improved by correcting the faults that were detected, adding new information and deleting existing items where needed (Van Aken et al., 2012).

Documents (secondary data collection)

During interviews relevant documents and archival records were asked for (Yin, 2014), in order to get additional information about the cases. Received documents were storyboards about the projects, speaker notes belonging to the storyboards for stakeholder meetings, implementations plans, and emails of external companies that were involved in the various projects.

Case descriptions

The case companies that have been selected are Starwood Hotels & Resorts, Apollo Hotels & Resorts, and citizenM. Below the five cases that have been selected and researched are described in short.

Case company 1: Starwood Hotels & Resorts the Netherlands

Starwood Hotels & Resorts Worldwide, Inc. currently has 4 hotels in the Netherlands: Sheraton Amsterdam Airport Hotel and Conference Center, Element Amsterdam, W Amsterdam and Hotel Des Indes, A Luxury Collection Hotel, The Hague. Until April 2015 it also managed Hotel Pulitzer, A Luxury Collection Hotel, in Amsterdam.

Case 1: Successful ‘Waiting is Fun’ project at Hotel Pulitzer

(15)

14

Case 2: Unsuccessful ‘Pre-arrival Experience’ project at Sheraton Amsterdam Airport Hotel

Sheraton Amsterdam Airport Hotel and Conference Center is located on Schiphol airport. It is the closest hotel to the terminals and train station with a 3-minute walk to the arrivals/departures hall. A threat was seen in the new Hilton hotel that opened next to the Sheraton end 2014 that was much bigger and newer than the Sheraton hotel, which could cause losing guests. The management team came together to discuss the possibilities to strengthen Sheraton‟s position in the competitive set. The brainstorming resulted in the decision to take advantage from its unique selling point (the location), with the idea to increase guest loyalty by creating a pre-arrival experience for guests at the airport before even entering the hotel. This project also started with a two-day fast track session, in which many opportunities were voiced. All possible options were divided among the project members, who investigated whether the options could be executed. Unfortunately, several ideas could not be developed because of rules and regulations of Schiphol airport; it was for example not possible to place a Sheraton information desk behind the customs. Later on, all hopes were put on developing an app in collaboration with an external party, which could virtually lead the guests from arrival at the gate to the hotel. The actual app development never happened, as it was technologically not feasible and because it would be an excessive investment. Eventually, the project was stopped by the project team, as the most valued options to fulfill the guests‟ needs were not feasible to be carried out.

Case company 2: Apollo Hotel Groningen

Apollo Hotel Groningen is a modern hotel located in the distinctive building La Liberté in the skyline of Groningen, which consists of two parts connected by a bridge. It opened the end of June 2015 and belongs to the Dutch hotel chain Apollo Hotels & Resorts.

Case 3: Successful development of the new service concept

(16)

15 months after the opening it could be seen that an organization was established that supports itself and that the concept was carried out as expected. The concept is seen as successful as both employee satisfaction and guest satisfaction are high. Besides, it also contributes to the hotel financially, as returning guests are already seen which increases the hotel‟s revenues.

Case 4: Unsuccessful signing project

The way in which the non-verbal guest communication – by means of signage – is executed, is a project performed at the Apollo Hotel Groningen that completely failed. The innovation process is carried out until the implementation of the new signs in the hotel, however the outcome is not as desired. The reason for implementing signage throughout the building became clear early in the pre-openings phase, as it a very complex building. Therefore, it was included in the specifications and conditions for the building that it had to be fitted out with signage. A signage company was approached prior to the opening, however they could only come for the first site inspection after the building was done and furnished. The inspection took place later than expected. First reason is that the hotel opened earlier than intended, and second, the date was set wrong by the signage company which caused the site inspection to be postponed. During the process, the communication between the signage company and the hotel was not optimal; it mostly went via email and telephone, since the signage company was not closely located. What is remarkable in this project is the fact that the signs have not been tested extensively. Each time proposals came in, the general manager and operational manager made a round through the building to evaluate the proposal, after which they gave feedback to the signage company. However, they were familiar with the company and should have let others test it in order to see whether it actually works. Moreover, the signage company did not provide proper recommendations and advise about the places and colors that were chosen. Eventually, when the signs were delivered, some were misprinted or even missing. Therefore the signage company had to return several times to do restorations and replacements. Eventually, two years after the opening of the hotel, the management is still not satisfied with the signage of the hotel as it had cost a lot of money and guests still experience issues finding their way around in the hotel.

Case company 3: citizenM

citizenM is a relatively new hotel chain, which opened its first hotel at Schiphol in 2008. The chain is named after modern travelers that are explorers, professionals and shoppers, who are called the Mobile Citizen of the World. citizenM aims to bring „affordable luxury‟ to all „mobile citizens‟ of the world by opening properties on prime locations in metropolitan cities and at major international airports.

Case 5: Successful eco-warrior project at citizenM Amsterdam

(17)

16 further. This project, called the „eco-warrior project‟ developed the possibility for guests to choose to not let their room be cleaned with the use of a door hanger, which looks similar to the known do-not-disturb sign. In return they get rewarded with a drink/meal for a fixed amount similar to what is being saved. For citizenM it is not about the saving money, but about supporting the environment. The idea was developed around the summer in 2016 for two reasons: Firstly, for a long time, citizenM had it on their agenda to do something extra for the environment in which they could involve the guests more, and secondly, a citizenM employee spotted the opportunity in a hotel in Londen to choose to not let the room be cleaned and get a drink in return. Colleagues of citizenM started questioning how they could contribute to the environment and reward guests for it, without having the intention to reduce costs. A project team was set up to get the project started. The team went through several phases in which many conversations and brainstorming sessions were held with internal stakeholders that were identified - the hotel team and the housekeeping partner. It was decided by the project team not to go live until all stakeholders were comfortable and aware of everything that had to do with the new service. In November 2016, the eco-warrior service went live as a pilot in citizenM Amsterdam. The pilot‟ performance was closely monitored by collecting feedback from the guests and in reviews posted online. In a weekly meeting the project team sat together with the stakeholders for sharing thoughts and feedback. Eventually, the pilot got extended due to the idea to do it somewhat differently for the official launch. Even though the pilot runs successfully, and it has not been the intention to change the project at first, the project team thinks that it can be done even better and decided to make a change. The goal became to make the guests aware that the entire amount that citizenM saves is going to be donated to a charity. The money will thus not be returned to the guests in the form of food and beverages anymore. Currently the project team is finalizing the plans identifying the charity and how exactly that is going to be performed in practice. The project is about to be launched as a new pilot of three months in the near future, but compared to the first pilot it will be implemented in all citizenM hotels. When the second pilot phase comes to an end, the project team has to come up with the decision on whether to officially implement the service in all hotels. Currently this is expected to become a success, as the pilot in citizenM Amsterdam runs successfully too.

Data analysis

(18)

17 For analysing the data, the method of Eisenhardt (1989) was followed. First a within-case analysis was performed for every single case. With help of the transcripts and supporting documents case descriptions were written. The activities found in the case descriptions were chronologically placed in a diagram and compared with the activities in the unsuccessful project. Thereafter, a cross-case analysis was performed; the insights of the cases were compared to each other, looking for similarities and differences, and pattern matching took place with the help of tables (Eisenhardt, 1989). Here, the notes that were made during the interviews were of some help. As the companies did not always use similar words for the same sort of activities, the insights that were obtained were continuously refined and overarching words were developed that cover a wide range of descriptions. The frame that has emerged from this cross-case analysis had to be compared to each case again, in order to determine if it successfully fits with the case data (Eisenhardt, 1989). Thereafter, the findings of this cross-case analysis were compared to what is already presented in the academic literature about NSD processes, and the differences and similarities have been discussed. Lastly, a new NSD process model is proposed based on the findings of this research.

Research Quality

(19)

18

Quality criteria Case study tactic Research phase

Controllability  Detailed description is present of the steps that

are followed to conduct this study

Research design, data collection, data analysis

Reliability  Use of a case-study protocol

 Triangulation: use of multiple instruments  Several respondents per case company  Interviews were performed at different

moments of the day and contact via email was never on the same time of the day

Data collection Data collection Data collection Data collection

Construct validity  Multiple sources of evidence

 Interviewees review interview summaries and case descriptions for comments and remarks.  restore, remove or add new information

Data collection Data collection / Data analysis

Internal validity  Pattern matching

 Explanation building: the emerged theory closely fits the collected data from the cases, thus is highly tied with evidence.

 The emergent theory is compared to existing literature

Data analysis Data analysis

Data analysis

External validity  Specified population

 Replication logic in multiple-case studies

Research design Research design

Table 3 - Controllability, Reliability and Validity tactics (Eisenhardt, 1989; Van Aken et al., 2012; Yin, 2014)

FINDINGS

(20)

19

The NSD process

The in-depth analysis of the activities that are performed in the investigated cases has resulted in a process model for NSD in the hospitality industry. The model is demonstrated in table 4 on the left and consists of 8 phases that demonstrate 16 aggregated activities in total.

‘What’ – Activities

The detailed activities that have been performed across the investigated cases are also demonstrated in table 4. On the basis of this table the activities are described per development phase.

Problem recognition

It can be seen in table 4 that four out of the five cases clearly started with problem recognition. In the first case a long lasting problem was recognized and it was decided that a structural solution was needed. Thereafter a problem analysis and assessment were performed. Case 4 also shows problem recognition; it was recognized before the opening that it is a difficult building in which the hotel would be located and that it needed some sort of signage. However, in cases 2 and 3 a foreseeable problem was acknowledged rather than a problem which already existed; in case 2 a threat was seen in the opening of the new hotel next door as it might cause losing guests, and in case 3 a problem was foreseen of not being able to differentiate oneself from others when missing a specific concept. Only case 5 did not start with a problem; instead it started with a wish of the company itself.

Idea generation

An idea was generated in all cases. For case 5 this was the starting point of the project. Most cases (1, 2, and 5) recognized an opportunity, which was subsequently assessed (cases 1, 2, 4 and 5). The assessment was the first activity for case 4 in the idea generation phase, as the solution for the recognized problem was very obvious. Only case 3 seems somewhat different from the other cases in this phase as it doesn‟t start with opportunity recognition, but with identifying guidelines given by the hotel chain within which the concept must be formed. The detailed activities hereafter (6-9) contribute to the concept generation of the service concept. Interesting to notice is that both cases of Apollo document the idea/concept that has been generated as the last step of this phase, whereas the other cases do not document their initial idea/concept.

Project planning

(21)

20 the stakeholders (citizenM). Starwood also made the decision to execute the projects with help of the 4D method at this point. Secondly, preparation activities for the first meeting came forward. In cases 1 and 2 the guests were included (detailed activity 16) by collecting data about their needs (case 1) and obtaining creative solutions/ideas (case 2). Detailed activities 14 and 17 show that the first meetings with the stakeholders or external partners were prepared. Noteworthy is that case 3 did not go through this phase at all; no project team was set up nor was a planning made or preparation done.

Define phase

Table 4 shows that all cases defined the goal of the project (detailed activity 18), which is the starting point of the define phase. It can be seen that Starwood performed more activities in cases 1 and 2 in this phase compared to the other companies. This is due to the fact that the 4D approach is followed, which included several methods to identify the scope of the project, the expected outcomes of guests and what opportunities would have the biggest impact. All these activities have been performed during the scheduled fast track. In case 5 no other activities were performed during this phase. Apollo however defined the practical execution of their concept in case 3, and the scope of the innovation in case 4 as the signing company measured the area for the innovation.

Discover phase

This phase includes generating ideas, performing feasibility and value assessments and developing an action plan. Again, this phase shows most activities being performed by Starwood. In both cases ideas were generated by means of brainstorming sessions and other idea generating tools. Interesting is that with the latter ideas were discovered to improve related internal processes that already existed, as an extension to the project (detailed activities 31-32). Case 2 performed more activities to gather information for creative ideas from different sources (detailed activities 26-29) due to the fact that the focus area of the project was largely outside the hotel. Generating ideas this extensively is hardly seen in the other cases; as the solutions in cases 3, 4 and 5 were generally obvious and already defined in the idea generation phase, there were no brainstorming activities needed here. Moreover, three out of the five cases (1, 2 and 5) discussed the idea/project feasibility and developed an action plan as final step in this phase. In case 4 feasibility of the project was not discussed, however Apollo‟s management explored potential other partners and considered switching, as the management was not satisfied with the project‟s quality and speed (thus considered less feasible at that point).

Develop phase

(22)

21

Table 4 – NSD process activities from multiple-case study research

Case #1 - Waiting is Fun (successful)

Case #2 - Pre-arrival experience (unsuccessful)

Case #3 - Development of service concept (successful)

Case #4 - Signing project (unsuccessful)

Phases Activities Detailed activities Detailed activities Detailed activities Detailed activities

1 1 Decision to pick up the problem (structural solution needed)

- Decision to pick-up foreseeable problem

-2 Problem analysis and

assessment

Problem analysis and assessment (threat of Hilton)

- Problem analysis and assessment

2 3 Opportunity recognition Opportunity recognition -

-4 Idea assessment Idea assessment - Idea assessment

5 - - Identify concept guidelines

-6 Concept generation Concept generation First concept generation Solution decision

7 - - Identify practical application

-8 - - First concept finalization

-9 - - Documentation of concept Documentation of solution

3 10 Set up a cross-functional team Set up a cross-functional team - Decide and contact external partner

11 Define roles & responsibilities Define roles & responsibilities - -12 Decide on method to execute

the project

Decide on method to execute the project

-

-13 Decision to schedule a fast track Decision to schedule a fast track - Schedule first meeting with partner

14 Preparation of fast track - - Preparation of first meeting 15 Front-line employee preparation

and engagement

- -

-16 Doing research: customer probing & data collection

Customer engagment in preliminary idea generation

-

-17 Identify internal stakeholders for fast track

Identify internal stakeholders for fast track

-

-4 _{5. Define goal} 18 Set up goal statement innovation opportunity

Set up goal statement innovation opportunity

Set up goal statement (putting a dot on the horizon)

Clarify goal to external partner 19 Define scope of the project and

exclude outcome expectations that are outside the scope

Define scope of the project and exclude outcome expectations that are outside the scope

-

-20 - Define customer touch points (focus areas)

- Define and measure innovation area (building)

21 Determine customer needs Determine customer needs - -22 Define outcome expectations Define outcome expectations - -23 Create outcome statements Create outcome statements Define execution of the concept

-24 - Prioritize outcome statements -

-25 Define biggest innovation opportunity

Define biggest innovation opportunity

-

-Opportunity recognition and analysis -Project planning Idea Generation

-Set up cross-functional team Define roles & responsibilities

-Schedule kick-off meeting

-Problem recognition 8. Develop and select solution concepts 3. Group formation

& project planning 1. Problem recognition 4. Preparation 2. Idea/Concept generation 6. Scope innovation opportunity 7. Create innovation opportunity Define

Apollo Hotels & Resorts

Starwood Hotels & Resorts

Case #5 - Eco-warrior project (successful)

citizenM

-Identify internal stakeholders for kick-off meeting

Clarify goal and reasoning in kick-off meeting

-Detailed activities

Idea assessment Define conditions for idea Concept generation

(23)

22

5 26 - Acquire external information -

-27 - Emerge in customer experience -

-28 - Assess customer ideas (voice of the customer)

-

-29 - Internal analysis Starwood Worldwide - -30 Brainstorming to generate creative ideas Brainstorming to generate creative ideas -

-31 Generate ideas with help of reverse thinking

Generate ideas with help of reverse thinking

-

-32 Perform analysis on existing internal process (SCAMPER)

Perform analysis on existing internal process (SCAMPER)

-

-33 Discuss idea feasibility Discuss idea feasibility - Explore other partners / consider switching 34 Value assessment of ideas Value assessment of ideas -

-35 - Identify external partners/stakeholders

-

-36 Convert ideas into actionplan Develop action plan -

-6 37 Creating services Execute action plan Develop solution (signs) &

adjustments 38 Involving employees by means

of meetings with them & update the sponsor

Contact and cooperate with external partners

- Send/receive proposal for solution

39 Evaluate and adapt solutions Evaluate solutions & revise action plan

- Evaluate solution & send feedback

40 - -

-41 - Decision to cease the project -

-13. Documentation 42 Documentation of the project Documentation of the project -

-14. Prepare staff for go-life of pilot

43 Training/workshop with front-line employees

- Front-line employee training

-7 44 - - Create open company culture

-45 - - Set up self-sufficient

organization

-46 - - Continuous evaluation

47 Experimenting / Testing - Experimenting / Testing - Experimenting /

Testing the service

48 - - Concept delivery recognition -

-49 Evaluating the pilot - Evaluate the pilot (formal concept evaluation)

- Evaluation of the pilot

50 Final changes/improvements to service

- Adjustments & Refinements - Adjustments /

refinements to service goal and concept & idea assessment

51 - - - - Identify & update all

stakeholders

52 - - - - Experimenting /

Testing the service

53 - - Adjustments & Refinements - Evaluation of the pilot

8 Launch 16. Launch 54 Launch - Service concept continuation Placement of solution (signs)

55 - - - Evaluate solution & refine

solution 56 - - - Restorations 11. Develop action plan

-15. Pilot testing & further educate service employees 10. Feasibility and value assessment 9. Generate ideas Demonstrate / Testing Develop Discover 12. Solution generation First pilot Second pilot

-Launch / service continuation (OR cease service)

-Update all stakeholders Documentation of all stakeholder effects

-Evaluation of stakeholder effects

Stakeholder meetings to identify impact & processes

-Identify necessities and develop action plan

-Identify purpose & fit with company values

Discuss and define project feasibility

(24)

-23 developed and therefore the project team decided to cease the project. The develop phase furthermore shows that the projects have been documented in most cases (1, 2, and 5), whereas in case 3 the concept was already documented during the idea generation phase. Case 4 does not show documentation activities. Finally, in most cases the entities prepared their staff for the go-life of the pilot as the last activity in this phase; cases 1 and 3 executed a front-line employee training/workshop, which would probably been done in case 2 as well if it was not ceased. Case 5 shows a similar activity: The project team sent around an update to all stakeholders which also indicated the go-life for the pilot. Only case 4 lacks this activity.

Demonstrate/Testing phase

This phase includes educating service employees and testing the new service. Educating service employees is seen in case 3, where the detailed activities 45-48 contribute to setting up a self-sufficient organization. This in turn contributes to the successful delivery of the new service. Cases 1, 3 and 5 performed a pilot test and the project team evaluated it. This highlighted the need for some final refinements to the services. Case 3 shows iterations as developing and testing was performed simultaneously during the pilot test; an organization is developed, as aforementioned, while demonstrating the service and at the same time being evaluated and refined. Also remarkable is that case 5 demonstrates a second pilot; despite a successful first pilot the project team decided to adjust the concept in order to increase the service innovation‟s value. Therefore, an idea assessment has been re-executed and another update to all stakeholders has been sent before the second pilot (detailed activities 51 and 52). Noteworthy is the fact that case 4 has not performed a pilot test at all.

Launch

The last phase is the launch, which in most cases (1, 3 and 5) turned out to be a continuation of the service. Only case 2 had no launch, as the project was stopped during the development phase. In case 1 it was found during the evaluation that the services were well developed prior to the pilot test and therefore these were simply extended. Case 3 shows a continuation of the service, as the management was satisfied with how the concept was carried out by the employees. Case 5 also shows a service continuation (or cease). The second pilot has to be evaluated and it has to be decided whether to continue the service in the citizenM properties or whether it will be ceased in some properties. Case 4 shows more activities in this last phase; an evaluation, refinements and several restorations because mistakes were made while placing the signs. Therefore, this phase took longer than in the other cases.

‘How’ – process formality

Different degrees of process formality are seen across the case companies.

(25)

24 documented in the form of a PowerPoint presentation, a storyboard (one-pager including main activities and findings) was developed and all documents (including the workshops, storyboard, implementation plan, and speaker notes) have been safely stored for potential future use in other Starwood projects. Thirdly, responsibilities are clearly defined depending on the project members‟ yellow belt, green belt or black belt degree and knowledge of the subject. Moreover, the projects include specialized roles such as project leaders (both on project- and operational level) and sponsors. Apollo shows a loose process formality structure in its NSD processes. Firstly, it shows a low degree of systematic behavior; no systematic procedures and rules govern the development processes, especially not in case 3. Case 4 shows it to some extent, because the external company followed its own regular procedure, however from the hotel‟s perspective no predefined procedure was followed. Secondly, both cases show documentation to some extent, as the concepts were documented at the start of the process, but later on in the process this practice was abandoned. Thirdly, responsibilities were not clearly defined in both cases as no project team was been set up in any form whatsoever. Lastly, citizenM shows a high process formality in case 5. Firstly, the degree of systematic behavior is quite high: A planning was made with a deadline for the pilot and everything that had to be done prior to the go-life. Besides, citizenM holds itself to some guiding principles in all their projects: The purpose and context of the project should be leading, all important stakeholders should be involved, and everything should be done with the guest in mind. Secondly, case 5 shows a high level of documentation: The effects for all stakeholders are clearly documented in the develop phase and during the pilot phase its success is tracked in spreadsheets. Thirdly, responsibilities are clearly assigned based upon knowledge and interest at the beginning of the project.

‘Who’ – cross-functional involvement

(26)

25

Successful NSD vs. unsuccessful NSD

Cases 1, 3 and 5 have established successful new service innovations. The companies generally used the same measurements for success to identify the service‟s success: Guest satisfaction, employee satisfaction and (financial) performance. Guest satisfaction is seen as the most important success factor by all three companies, and is measured via feedback in practice, online reviews and the GEI score. Employee satisfaction is also measured by Starwood and Apollo. As for these specific cases the front-line employees had to deliver the service to the guests, thus their satisfaction was extremely important too. The financial performance was also measured, but less important, especially for citizenM as it was not their goal to reduce costs with this particular project but rather do something for the environment. Instead, the amount of service users was deemed more important, as guests can choose to use it or not. Cases 2 and 4 were unsuccessful cases. Case 2 was labeled unsuccessful because it was aborted during the NSD process as the service was not able to be developed at that time. Case 4 is labeled unsuccessful because it did not result in the desired result; guests satisfaction concerning the issue is low, and the hotel‟s management is unsatisfied about it as it has cost a lot of money, yet produces no new revenue.

Key differences between successful- and unsuccessful projects

The respondents of the interviews referred to several factors that made the successful cases to a success: The fast track, the sponsor role, the well prepared and performed pilot tests and good communication to employees and internal stakeholders.

The fast track

In the Starwood cases, the fast track was designed to get the project started; the project group had an intensive workshop for two days in which they would not be disrupted by the day-to-day hotel operation. The first phases of the 4D approach were walked through in which well thought out solutions were generated for the recognized issues. The fast track‟s output was an action plan so that everyone knew what had to be done. The fast track was also performed in the case 2, which ensured a good start for that project too, even though it failed at a later stage. A similar observation can be made in case 5 of citizenM where a kick-off session was scheduled with the internal stakeholders. This ensured each key person involved was informed and updated, so that everyone could contribute the knowledge that was required and share concerns that had to be clearly thought through.

Sponsor role

(27)

26 during the process. In this case the sponsor role was not actively and enthusiastically taken. Therefore, the project did not have the ultimate facilitation as the first researched project had enjoyed.

Pilot tests

Thirdly, the pilot tests have been pointed out as important elements for success. Table 4 shows that cases 1, 3 and 5 (the successful cases) all performed a pilot test. As all pilot tests were evaluated some last changes, adjustments and refinements were done before they officially launched (or simply continued, as previously mentioned). Case 2 did not make it to the pilot test, and case 4 shows to have skipped this element. Case 4 demonstrates some activities in the last phase (launch) that the other successful cases performed in the demonstrate phase where the pilot tests were carried out.

Employee involvement

Lastly, good communication to employees and internal stakeholders is seen as important. The detailed activities 15, 38 and 43 (see table 4) show that the employees were involved in the process and well prepared for data collection as well as the pilot test. Detailed activity 43, were front-line employees were prepared for the pilot shows in all the successful cases. It was not observed in case 4.

Clarification with life-cycle and teleological perspective

The reason why the two cases performed by Starwood show way more activities in the define phase and discover phase, and have the highest degree in process formality compared with the other cases, can be explained by means of the life-cycle perspective. The sequence of activities is predefined since the 4D approach was chosen as method for this project, which clearly demonstrates the life-cycle perspective: “the developing entity has within it an underlying form, logic, program, or code that

regulates the process of change and moves the entity from a given point of departure toward a subsequent end that is prefigured in the present state” (Van de Ven & Poole, 1995, p. 515). It

followed a sequence of NSD stages that is logically irreversible, linear and predictable such as Cooper‟s (2008) stage-gate process, following a problem-solving approach (Van Aken et al., 2012) with the methods applied during the process.

The large amount of empty cells in case 3, where it can be observed that the project planning, discover and develop phase are largely skipped is striking. During the demonstrate/testing phase development activities are seen; the development of a self-sufficient organization which is needed to perform and deliver the service concept. Iterations are also seen here. This can be explained by the teleological perspective that this case seems to have taken, which proposes that the progression through activities in the NSD process is guided by a goal (Van de Venn & Poole, 1995). This is also the case in case 3 where the goal has been set early in the process (activity 18) and where the activities that are performed later (23 and 43-50) have all been performed in order to reach that end state. “One of the

(28)

27

everyone is allowed to make decisions and has responsibility will have to start with educating people in order to make them feel that they really have those freedoms and especially which freedoms they don’t have.” (Operations manager, Apollo) This shows that it was needed to create a self-sufficient

organization in which people are aware of the concept before they are able to behave accordingly. Also the striking occurrence of a second pilot in case 5 is explainable by means of the teleological perspective; performed activities and the desired end state are not fixed but can be adapted during the process (Van de Ven & Poole, 1995), which explains the change of the project‟s goal and adjustment to the service causing a second pilot to be necessary. Since the renewed service concept had to be assessed and developed prior to go-life of the second pilot, the first pilot was extended and thus took longer than the initial three months. “When we started the pilot, even when we started this project, it

was not per definition the intention to change the project” (Operations Specialist, citizenM).

DISCUSSION

The literature performed on the process of NSD is not very recent and no accepted model for NSD has yet been developed (Biemans et al., 2016). De Jong and Vermeulen (2003) expressed the need for alternative descriptions of the NSD process, and Biemans et al. (2016) pointed at the benefit of an in-depth case study research. This research responded to the need for further exploration by using this research design. Firstly, an answer is found on how the NSD process looks like from idea generation to launch. Secondly, critical success factors in the NSD process are addressed. And thirdly, this research elaborates on the recent discussion raised by Biemans et al. (2017) of how NSD processes are performed when applying the life-cycle and teleological perspective. This chapter demonstrates a discussion about this study‟s results in comparison to the (service) innovation literature.

The NSD process

The process model that is found on the basis of the in-depth analysis of activities is showed in figure 1. This model consists of 8 phases, which demonstrates 19 aggregated activities in total.

Figure 1 – NSD process model

4. Preparation 1. Problem

recognition 1. Problem

recognition 15. Pilot testing &

further educate service employees 12. Solution generation 13. Documentation 16. Launch 2. Idea/Concept

generation 5. Define goal

9. Generate ideas 3. Group

formation &

project planning 6. Scope innovation opportunity 8. Develop and select solution concepts 10. Feasibility and value assessment 11. Develop action plan 14. Prepare staff for go-life of pilot 7. Create innovation opportunity 8. Launch 2. Idea generation 3. Project planning

(29)

28 When comparing this discovered model to the NSD models proposed in the literature, it is observed that some of these eight development stages are the same as reported in the other models of Bowers (1987, 1989), Scheuing and Johnson (1989), Avlonitis (2001), and Alam & Perry (2002): idea generation, (pilot) testing and launching. However, this model also reveals some differences. A new phase compared to the aforementioned models is the first phase: Problem recognition. Four out of the five investigated cases showed that before an idea for service innovation could be generated, first some sort of problem or foreseeable problem had to be recognized as motive for idea generation. Therefore, this has been added as a separate phase prior to idea generation. The fifth case study, however, started in the idea generation phase as there was not a (foreseeable) problem recognized, but instead the company possessed a wish for a new service due to other reasons. This corresponds to some extent with the phases „formulation of new service objectives‟ (Scheuing & Johnson, 1989) and „strategic planning‟ (Alam & Perry, 2002), which is in this model merged in the idea generation phase. This shows that the problem recognition phase can be skipped in cases where the reason to develop new service innovations is not due to a problem.

The third phase (project planning) includes group formation- and project planning activities, such as cross-functional team formation, division of roles and responsibilities and scheduling first meetings, and preparation activities, such as data collection and employee involvement. In the literature the success of cross-functional involvement has already been mentioned several times, however so far only the model of Alam and Perry (2002) included cross-functional team formation as an activity. Avlonitis et al. (2001) examined it as a block of variables belonging to the NSD process, but did not include the formation as an activity to be performed. This research considers this an important activity as part of the project planning phase, given that the majority of cases showed this activity and that the respondents indicated that it has contributed to a certain extent to the service innovation‟ success.

(30)

29 personnel training is seen as separate phases in previous models (Scheuing & Johnson, 1989; Alam & Perry, 2002), this model includes it as last activity in the develop phase, in order to prepare staff for the pilot‟s go-life.

Besides the often discussed importance of the idea generation phase, Biemans et al. (2017) found that the testing phase is also crucial for successful new service development. This research confirms this finding, as the project in case 4 turned out to be unsuccessful due to, among other things, missing testing activities prior to the launch. It is agreed with Biemans et al. (2017) that NSD managers should indeed execute test with frontline employees as well as customers and that these tests should be evaluated very carefully prior to the implementation of the service. This research shows that the successful cases did this very well, and that unsuccessful case 4 completely skipped these activities. In addition, a crucial activity has been found in this research in the testing phase, namely the activity of educating service employees. This differs from the literature on NPD where this is intertwined in the launch. This difference seems rather logical, as employees have to work with the new service already during the pilot, whereas for new products this is not necessarily the case.

The life-cycle and teleological perspective

(31)

30

Theoretical contributions

This research contributes to the literature in several ways. Firstly, it has answered to the need for more alternative descriptions of the NSD process (De Jong & Vermeulen, 2003) by investigating the NSD process in pure service environments and creating a new model on that basis. The discovered phases of the process and the activities therein are described and turned out to be somewhat different compared to the previously developed NSD models. Secondly, as recommended by Biemans et al. (2017) this research went deeper into the individual activities and looked at their relevance. Thirdly, since an accepted model of the NSD process is currently missing (Biemans et al., 2016) this model contributes to the literature by providing a potential model to be evaluated and accepted. Furthermore, this research gives insights into critical factors that make the difference between successful and unsuccessful NSD outcomes; the importance of testing activities is proven and other contributing factors such as a fast track, sponsor roles and proper employee involvement are discussed. Moreover, this research answers the suggestion given by Biemans et al. (2017) by incorporating a more comprehensive usage of the teleological perspective in order to discover characteristics of a fluid NSD process; the order and importance in which phases are performed as well as the use of feedback loops and iterations are clearly demonstrated in this research.

Managerial implications

The findings of this paper also have some managerial implications for NSD managers. First of all, NSD managers should see the problem recognition phase as a potential phase to be executed prior to the idea generation phase. This phase only applies when a problem or foreseeable problem is recognized; then a proper problem analysis and assessment should be performed before generating an idea/concept. When the development process starts for other reasons, it can directly start in the idea generation phase. Secondly, NSD managers should notice the ability to skip the first activity in the discover phase; generating ideas. This implies that NSD managers should evaluate whether the service that is going to be developed is rather obvious and already clearly defined in the idea generation phase, or whether brainstorming sessions and activities alike are needed to discover new creative ideas. Besides these practical implications of the newly developed process model, the NSD managers should be aware of the importance to include front-line employees and customers throughout the NSD process. Involving the front-line employees throughout the development process creates understanding and willingness to participate where needed.

New Service Development Process: