Challenges of Implementing Artifical Intelligence Related Services. What Enhancing and Inhibiting Factors Need to be Addressed by Consultancies?

Table of Content

Abstract

Introduction

Literature review
Capabilities for knowledge-based service innovations
Competence enhancing and disruptive technologies

Methods
Research design
Sampling approach
Data collection
Data analysis

Findings
Relevance perception of Artificial Intelligence
Client projects supported by Artificial Intelligence
Determinants for enhancing AI capability development
Determinants for inhibiting AI capability development
Critical remarks on internal AI capability development

Discussion
Theoretical contribution
Practical implications

Limitations and future research

REFERENCES

What are enhancing and inhibiting factors for consultancies to implement Artificial Intelligence related services?

Abstract

Artificial Intelligence (AI) based services innovation gains ground in consulting services because the technology allows due to its maturity now substantial benefits. Following service innovation theory, research evidence suggests that Artificial Intelligence is transforming the consulting industry, but there is disagreement about what this entails and in how far AI is a normal service innovation as others are - or whether it is disruptive to existing competencies and thus business models of consultancies. I conducted semi-structured interviews with 12 consultants from 10 different consultancies. As the main part of my research I found enhancing as well as inhibiting factors for consultancies on their way to implement Artificial Intelligence related services. I compared those to the relevant literature about service innovation capabilities against the background of the discussion in the literature whether the service innovation of AI is competence enhancing or competence disruptive. It turned out that, contrary to prior assumptions, the factors did not differ significantly from those of other service innovations. Exceptions from that were the factors of (co-)developing an applicable use case together with the client as an enhancer and the factor of lacking data quality (from the client) as an inhibitor. For this reason, these factors on both sides represent the only Artificial Intelligence specific service innovation capabilities for consultancies.

Introduction

The 4th industrial revolution is driven by information and communications technology. digitization, robotics, Artificial Intelligence and machine learning are relocating more decision-making from human workers to machines(Syam & Sharma, 2018). New intellectual property leads to new tool kits and frameworks, which in turn leads to further automation and technology products (Christensen, Wang, & van Bever, 2013). The level of intelligence of smart machines is increasing (Davenport & Kirby, 2016). The general trend is towards greater autonomy in decision making - from machines that require a highly structured data and decision context to those capable of deciphering a more complex context (Davenport & Kirby, 2016). The maturing of Artificial Intelligence offers possibilities to complement and optimize the existing service portfolios of consultancies(Nissen, 2017).Automated low-cost consulting solutions based on AI can open up whole new segments of clients that would otherwise never approach a consulting firm(2017). Software and technology-based analytics and tools that can be embedded at a client, have the potential to provide ongoing engagement outside the traditional project-based model (Christensen, Wang, & van Bever, 2013). According to Nissen(2017), business models of consultancies can be significantly modified if data are used to digitalize the service provision and enable automated consulting approaches that clients use autonomously. Artificial intelligence requires both substantial data and processing power, and both are available in large quantities today(Kokina & Davenport, 2017). Big data and analytical applications, in particular when combined with Artificial Intelligence and or machine learning, provide capabilities that raise consultants’ productivity and quality of results(Nissen, 2017).

The potential is immense, however according to(McKinsey Global Institute, 2016) most companies are capturing only a fraction of the potential value from data and analytics. This must also be considered against the background of the circumstances that the network effects of digital platforms are creating a winner-take-most dynamic in some markets (McKinsey Global Institute, 2016). Research evidence on service innovation uniformly confirms the question whether Artificial Intelligence is transforming the consulting industry. However, a debate is prevalent in the existing literature whether the service innovation of Artificial Intelligence is from the view of consultancies competence disruptive or not. Nissen(2017) listed Artificial Intelligence as one of the trends that may be interpreted in the context of a potential digital disruption of the consulting industry. As technology-based consulting solutions will gain ground business and delivery models of consulting have to be redesigned and partly reinvented within the next years (Nissen, 2017). Nissen (2017) even goes so far, to call AI a ‘key issue with growing relevance’.

Christensen et al.,(2013) stated that scores of start-ups and some incumbents are also exploring the possibility of using predictive technology and big data analytics to deliver value far faster than any traditional consulting team ever could. Though the full effects of disruption have yet to hit consulting, their observation suggests that it is just a matter of time(Christensen, Wang, & van Bever, 2013). They therefore argue in line with literature that sees Artificial Intelligence and related innovations such as big data analytics in the context of a disruptive service innovation. However they also relativize and stated that we are still early in the story of consulting’s disruption and that no one can say for sure what will happen (Christensen, Wang, & van Bever, 2013).

Other researchers see it differently and assign the service innovation of AI instead in a competence-enhancing context. As one of these, Sion (2018) concluded that Artificial Intelligence systems augment employees’ skills and team up with them to attain substantial output gains. This is in line with Davenport and Kirby (2016) who spoke in the course of their research with managers and found out that they do not expect machines to displace knowledge workers anytime soon instead they expect computing technology to augment rather than replace the work of humans.

The question that arises from this debate is who is right and consequently whether it requires new capabilities or not. In order to find an answer to this question, one would have to look at literature that sheds light on AI against the background of the differences between competence enhancing and competence disruptive innovations. The nascent state of literature underlines the relevance and importance of my research investigation. To close the literature gap, whether the innovation of AI is actually so different from other service innovations or just a service innovation comparable to others, I want to look at the competence development approach of consultancies for Artificial Intelligence. Since there is no scientific research existing and therefore a clear literature gap existing how consultancies can build up these AI competences, in my master thesis I address the research question “What are enhancing and inhibiting factors for consultancies to implement Artificial Intelligence related services?” To address the above-mentioned discussion of current research, whether AI is an enhancing or a disruptive innovation I compare the answers to the research question with existing literature on service innovation. I aim to show whether AI requires a different innovation process than other service innovations. By conducting semi-structured interviews with representatives of consultancies and analysing their experiences, I offer the following contributions to the literature on service innovation of AI.

First, I give an overview on the general perception of the importance of Artificial Intelligence competencies for consultancies. I show the different types of AI related services that are deliverable to clients based on Davenport and Kirby’s (2016) conceptualization. This ensures a better understanding.

Second, thanks to my exploratory research design, I am able to provide a rich description of enhancing and inhibiting factors for consultancies to implement Artificial Intelligence related services. The representatives in the interviews name specific factors for their organization, but most of them overlap with those in the literature for other service innovations. Especially for the big players in the consulting market, it turns out that they all have competences in AI and follow the technological trend. Some of them have competences that are at a very high level of what is possible today.

Third, I shed light on the discussion of current research, whether AI is a competence enhancing or a disruptive innovation. I provide an answer to the question whether it does really require new capabilities to set up the service innovation of AI or not as I demonstrate in depth that the determining factors do not differ significantly compared to other service innovations. Exceptions from that were the factors of (co-)developing an applicable use cases together with clients as an enhancer and the factor of lacking data quality (from the client) as an inhibitor. For this reason, these factors on both sides represent the only Artificial Intelligence specific service innovation capabilities I could find for consultancies.

The paper is structured as follows, firstly, the literature review addresses the theoretical background of this study by discussing literature on service innovation capabilities. Hereby, in particular, the factors that are associated to contribute as enhancing factors. Secondly, the methods section outlines my approach of the research design, data collection and data analysis. Thirdly, in the results section, the results of the study are presented, compared with the existing literature and placed in the discussion of the existing literature. Finally, in the discussion section, the results, their implications for service innovation theory and consultancies in practise are discussed. The research concludes with its limitations and the potential areas for future research.

Literature review

Capabilities for knowledge-based service innovations

Constantly developing new knowledge products is regarded as crucial in theory for suppliers to maintain demand for their services(Heusinkveld & Benders, 2005), (Huczynski, 1993). To develop knowledge-based innovations, the theory provides various approaches. In this section I want to give an overview what the literature has already said about consultancies to build up new capabilities.

Abbildung in dieser Leseprobe nicht enthalten

Table 1: Literature on capability enhancing factors.

Anand, Gardner and Morris (2007) have specified capabilites for knowledge-based innovation for management consultancies. They identified four critical generative elements: socialized agency, differentiated expertise, defensible turf, and organizational support(Anand, Gardner, & Morris, 2007). These elements must be combined in specific pathways for knowledge-based innovative structures to emerge and embed(Anand, Gardner, & Morris, 2007). These pathways can basically be distinguished into a bottom-up context on the one hand, where organizational support in the form of political sponsorship and access to firmwide resources help seed new knowledge structures(Anand, Gardner, & Morris, 2007). And on the other hand in a top-down context with direct intervention through goal setting and deployment of skilled or formally powerful people. Anand et al. (2007) have focused on individuals' agency as the starting point for innovation in knowledge-based organizations. But it is also widely acknowledged that innovations can result from interactions with clients whose needs indicate market demand (Von Hippel, 1986). Also according to Chesbrough and Schwartz (2007) one important mechanism for innovating a business model is by establishing co-development relationships with external parties. Those are considered as increasingly effective means of innovating the business model to improve innovation effectiveness(Chesbrough & Schwartz, 2007).

Den Hertog, van der Aa, and de Jong (2010) identified a set of dynamic capabilities for managing service innovation. Regarding the development of capabilities of innovation Kalali and Heidari(2016) showed that under changing environmental conditions, dynamic capabilities of consultancies can become a source of competitive advantage. Den Hertog, van der Aa, and de Jong(2010) hypothesized that successful service innovators outperform their competitors in at least some of their defined capabilities. Those six capabilities are: ‘signalling user needs and technological options’, ‘conceptualising’, ‘(un-)bundling capability’, ‘co-producing and orchestrating’, ‘scaling and stretching’ as well as ‘learning and adapting’. With this concept, they differ from Anand et al.(2007) in the sense that they involve external components. According to Den Hertog et al. (2010) the majority of these services propositions are co-created by the client and the provider. With the ‘signalling user needs and technological options’ Den Hertog et al. (2010) describe an intelligence function which they labelled as the capability to see dominant trends, unmet needs and promising technological options for new service configuration. ‘Conceptualisation’ is typically of importance in service innovation since on the one side, its predominantly conceptual nature makes it difficult for a customer to assess beforehand what will be experienced and what will be delivered(Parasuraman, Zeithaml, & Berry, 1985) and on the other side, its highly interactive or shared process character(Alam, 2002); (Magnussen, Matthing, & Kristensson, 2003). With the unbundling capability services can be stripped down to their bare essentials and highly specialised services created that are very similar and can therefore be standardized to a certain extent (Den Hertog, et al., 2010). The fourth capability ‘Co-producing and orchestrating’ is of considerably importance(Den Hertog, Van der Aa, & De Jong, 2010) since service innovations can be hard to introduce on a large-scale in a uniform way due to their intangible character, a human component which is hard to standardize, and their cultural dependency(Lyons, Chatman, & Joyce, 2007). At the same time, customers do expect to receive service in a similar fashion therefore, the capability of ‘scaling and stretching’ is considered as important by Den Hertog et al. (2010). And as a sixth the capability to deliberately learn from the way service innovation is managed currently and subsequently adapt the overall service innovation process is listed by Den Hertog et al. (2010).

Teece (2007) stated that service innovators are expected to be able to engage in alliances and networks as well as they have be able to manage and orchestrate these various coalitions. Mahmood, Zhu and Zajac (2011) examined how firms’ multiplex network ties in business groups represent one important source of capability acquisition.

Excellent companies invest and nurture in capability development, from which they execute effective innovation processes, leading to innovations in new product, services and processes, and superior business performance results (Lawson & Samson, 2001). Lawson and Samson (2001) named seven considerable factors for capability development which are: vision and strategy, harnessing the competence base, organisational intelligence, creativity and idea management, organisational structures and systems, culture and climate, and management of technology. Chen and Huang (2009) stated that human resource practices are the primary means by which firms can influence and shape the skills, attitudes, and behaviour of individuals to do their work and thus achieve organizational goals.

Competence enhancing and disruptive technologies

Prior research by Tushman and Anderson(1986) on technological changes which they divided between competence enhancing and competence destroying shows the patterns that while competence-destroying discontinuities are initiated by new firms and are associated with increased environmental turbulence, competence-enhancing discontinuities are initiated by existing firms and are associated with decreased environmental turbulence. These effects decrease over successive discontinuities and firms that are able to initiate major technological changes grow more rapidly than other firms (Tushman & Anderson, 1986). Against this background, the question that arises subsequently then is, where the service innovation of AI in this context is and whether AI-based services are then competence enhancing or competence disruptive?

Sion (2018) researched, whether cognitively enhanced machines will decrease and eliminate tasks from human workers through automation, and came to the conclusion that Artificial Intelligence systems augment employees’ skills and team up with them to attain substantial output gains. Firms that develop cutting-edge manners for machines to reinforce employees will be the ground breakers of their sectors (Daugherty & Wilson, 2018). In contrast, organizations that employ machines simply to substitute human workers will in the long run stall(Sion, 2018). This is in line with Davenport and Kirby (2016) who spoke in the course of their research with managers and found out that they do not expect machines to displace knowledge workers anytime soon - they expect computing technology to augment rather than replace the work of humans.

Nissen(2017) listed Artificial Intelligence as one of the trends that may be interpreted in the context of a potential digital disruption of the consulting industry. Nissen (2017) even goes so far, to call AI a ‘key issue with growing relevance’. He concludes that technology-based consulting solutions will gain ground (Nissen, 2017) and can therefore threaten existing competences. Particularly concerned with disruption in consulting, he lists 'analytics for decision making' and 'automation of routine tasks'(Nissen, 2017).

The state of prior theory in this field that can be categorized as nascent due to the novelty of the subject and recent emergence of relevantly mature AI underlines the relevance and importance of my research investigation. This topic is important since the literature agrees that business and delivery models of the consulting industry have to be redesigned and partly reinvented within the next years(Nissen, 2017). The potential of big data and analytical applications is too immense to ignore it. In particular when combined with Artificial Intelligence/machine learning, it provides capabilities that raise consultants’ productivity and quality of results (Christensen, Wang, & van Bever, 2013)(Nissen, 2017).

If it turns out that with the usage of Artificial Intelligence, (possibly if this is even more mature), existing business models lose relevance, a non-development of capabilities can have devastating consequences. Consultancies with AI competencies could use these for customer projects and also leverage synergies from previous AI related projects as the number of projects increases. Consultancies that have no such AI competences might no longer be able to offer client requests in the appropriate quality, speed and cost level and are in danger of gradually disappearing from the market. Winners will be differentiated from losers by their understanding of the evolving pressures on their clients and by their ability to bring clarity and skill to fulfilling clients’ new requirements (Christensen, Wang, & van Bever, 2013).

As a consequence, in my master thesis I address the research question: “What are enhancing and inhibiting factors for consultancies to implement Artificial Intelligence related services?”

Since there is a clear literature gap existing about which internal factors determine whether consultancies build up capabilities in AI, I want to close this with my contribution. To address the above-mentioned discussion of current research, whether AI is an enhancing or a disruptive innovation, I compare my answers to the research question with existing literature on service innovation and aim to show whether AI requires a different innovation process than other service innovations.

The following sections outline the rationale for the chosen method of data collection and analysis.

Methods

Research design

This thesis has a qualitative research design and seeks to offer a rich description of enhancing and inhibiting factors for consultancies to implement Artificial Intelligence related services. This type of research design is appropriate since the state of present theory and research on the factors that determine whether consultancies build up capabilities in Artificial Intelligence can be categorised as nascent(Edmondson & McManus, 2007). Therefore, I do not aim for statistical but theoretical generalizability. Given the rather minor developed nature of the research field and the open-ended question, a grounded theory approach whereby theory is ‘grounded’ in and developed out of the data (Gioia, Corley & Hamilton, 2013; Glaser & Strauss, 1967) and provides the compulsory ‘methodological fit’ (Edmondson & McManus, 2007). This grounded theory approach is appropriate since this field has not been extensively researched so far. This form of qualitative research design allowed me as the inquirer to generate a general explanation of which factors are enhanceing and which inhibiting for competences of Artificial Intelligence related services, shaped by the views of a large number of participants (Creswell, 2017; Strauss & Corbin, 1998) and to gain valuable insights in a relatively underexplored area.

To develop new theory, I relied on theoretical sampling (Eisenhardt & Graebner, 2007). I selected service innovation as a research context since Arificial Intelligence in consulting is primarily a service innovation.

[...]