Design thinking for innovation: context factors, process, and outcomes

Abstract

Purpose

Design thinking has become an omnipresent process to foster innovativeness in various fields. Due to its popularity in both practice and theory, the number of publications has been growing rapidly. The authors aim to develop a research framework that reflects the current state of research and allows for the identification of research gaps.

Design/methodology/approach

The authors conduct a systematic literature review based on 164 scholarly articles on design thinking.

Findings

This study proposes a framework, which identifies individual and organizational context factors, the stages of a typical design thinking process with its underlying principles and tools, and the individual as well as organizational outcomes of a design thinking project.

Originality/value

Whereas previous reviews focused on particular aspects of design thinking, such as its characteristics, the organizational culture as a context factor or its role on new product development, the authors provide a holistic overview of the current state of research.

Citation

Rösch, N., Tiberius, V. and Kraus, S. (2023), "Design thinking for innovation: context factors, process, and outcomes", European Journal of Innovation Management, Vol. 26 No. 7, pp. 160-176. https://doi.org/10.1108/EJIM-03-2022-0164

1. Introduction

Design thinking (DT) is an established process used in organizations, which aims to solve problems and promote innovation (Brown, 2008). By its creative und intuitive nature, DT can be distinguished from other processes, which are purely analytical (Mansoori and Lackéus, 2020; Nakata, 2020). Advantages for product or service innovations are promised through DT by a strong focus on the users' needs (Brown, 2008). Immersion in the user situation is considered to enable the discovery of (future) user needs. Other benefits include better decision-making by reducing cognitive biases (Liedtka, 2015), promoting learning effects (Beckman and Barry, 2007), and transforming the organizational culture toward innovation (Elsbach and Stigliani, 2018; Kolko, 2015). In the long term, case studies demonstrate that competitive advantages can be generated by applying DT practices (Appleyard et al., 2020).

Although design research dates back to the 1960s, the notion of applying design principles in the business context is still relatively young (Johansson-Sköldberg et al., 2013). Consequently, the emphasis of research has long been on defining characteristics to link the realm of design with business and management (Carlgren et al., 2016b; Micheli et al., 2019). More research has then been conducted on the possible uses, conditions, and effects in recent years (Chouki et al., 2021). For a long time, findings were primarily based on case studies describing practical applications (Holloway, 2009; Liedtka, 2015). Recently, researchers have started to use quantitative methodologies to measure the impact of DT (Suci et al., 2021; Nakata and Hwang, 2020; Nagaraj et al., 2020).

Due to its popularity in practice, scholarly attention on DT has increased over the years, leading to a growing number of publications on DT. However, the wide literature landscape can be characterized as complex and fragmented. As a consequence, we aim to provide an overview of the DT field and propose a research framework. For this purpose, we conduct a systematic literature review.

Our review complements previous reviews by a more holistic approach. Previous reviews focused on the characteristics of DT (Micheli et al., 2019), on DT and organizational culture (Elsbach and Stigliani, 2018), and recently on DT and new product development (De Paula et al., 2022).

Our review contributes to the DT literature by providing a holistic and systematic overview of DT as an established problem-solving and innovation process in organizations. In particular, we propose (1) an umbrella definition, which highlights the focal cornerstones of this specific process, and (2) a research framework, which organizes the findings of previous and future research. The framework comprises context factors, a basic three-stage process model unifying previous process conceptualizations, and typical outcomes of DT processes.

2. Methodology

2.1 Sample

This paper aims to provide an overview of the current findings of DT by analyzing the current literature on this topic. Definitions, characteristics, influencing factors, and impacts are elaborated based on the literature to create a research framework for DT. For this purpose, we conduct a systematic literature review (Kraus et al., 2020, 2022; Linnenluecke et al., 2020; Tranfield et al., 2003) (Figure 1).

The literature sample was retrieved from the databases Web of Science (WoS) and Scopus. In particular, the search was focused on publications with “design thinking” in the title to ensure that they deal with DT at their core and not as a side-aspect. The search yielded 794 documents on the WoS and 5,036 on Scopus, which shows that DT can be considered as a mature research field (Kraus et al., 2020).

Due to our thematic focus, we limited the search to the categories “business” and “management”. Following the recommendation by Kraus et al. (2020), we included research articles only. Furthermore, only articles written in English, the standard scholarly language, were considered. In terms of time, all texts published up until August 27th, 2021 are included in the review. After applying these filters, the number of articles decreased to 129 papers from WoS and 184 papers from Scopus.

We then further reduced the literature sample by screening the articles for their thematic relevance and minimum quality. Different quality thresholds can be applied, such as using journal rankings or minimum average citation rates of journals (Kraus et al., 2020) or minimum citation numbers of the articles in the sample, as often used in bibliometric reviews (Deyanova et al., 2022; Glinyanova et al., 2021; Tiberius et al., 2021). Considering the sample size and aiming for a rather broad overview of the field, we decided to apply a not too strict threshold. In particular, we removed articles published in journals with an Impact Factor or CiteScore smaller than one, which means that each article published in a journal had to be cited at least once, on average, showing a minimum relevance of the journal. In the next step, the merged dataset was checked for duplicates (Linnenluecke et al., 2020), removing another 71 documents. The final dataset contains 164 articles.

2.2 Analysis

As suggested by Kraus et al. (2020), we synthesized the data in a concept-centric manner. For this purpose, we read the titles and abstracts to obtain an initial overview. Subsequently, we read the texts completely for a deeper analysis (Denyer and Tranfield, 2009). We used Ma et al. (2019)'s model for our research framework structure, consisting of context factors, process, and outcome, to guide our review, and to categorize our findings. We added definitions as a further basic category. We used Excel to document the findings from the data analysis. In this process, we summarized the main statements of the articles in tabular form. Finally, we searched for relationships between the found insights and compiled them into a framework, which was then used to identify research gaps for future research.

3. Results

3.1 Defining design thinking

Today's understanding of DT originates from early design research in the 1960s (Elsbach and Stigliani, 2018), which aimed to disentangle complex unstructured problems into smaller, well-defined issues to develop better solutions (Beckman and Barry, 2007). Johansson-Sköldberg et al. (2013) distinguish between designerly thinking, i.e. the practices and competencies of designers applied within the design field, and design thinking, where non-designers adopt these design practices and competencies to solve particular problems. In this regard, Brown (2008) made a push toward the current understanding of DT in a management context by publishing his experiences with the consulting firm IDEO. According to this understanding, DT is a human-centered process that combines design methods with a business view (Brown, 2008; Henseler et al., 2021). Martin (2010) describes this as the balanced blending of analytical and intuitive thinking, which is expected to lead to a competitive advantage.

Due to different foci and perspectives, a uniform definition is lacking (Liedtka, 2015; Nakata and Hwang, 2020). Several definitions are listed in Table 1. The definitions show a wide spectrum of conceptualizations of DT. Some see DT as a discipline (Brown, 2008), an approach (Elsbach and Stigliani, 2018), attitudes/principles (Kolko, 2015; Shapira et al., 2017), thinking modes (Martin, 2010), a process (Beckman and Barry, 2007; Beverland et al., 2015; Glen et al., 2014; Liedtka, 2015; Shapira et al., 2017) or the application of methods (Seidel and Fixson, 2013). Whereas no understanding can be clearly declared as wrong, the different authors look at DT from different perspectives and with different foci. While the process perspective is predominant, the other conceptualizations should also be taken into account. To consolidate the different views, we propose the following definition:

Design Thinking is an iterative problem-solving and innovation process in organizations, which is based on specific principles (such as a focus on user needs, multidisciplinarity, etc.) and uses specific methods (such as creative thinking, visualization, experimentation, etc.).

3.2 Design thinking process

Several types of structures of the DT process have been proposed in the literature ranging from three to six stages (Table 2). However, the basic idea of the different process models is the same. Models with more stages only show a finer subdivision. Three basic stages can be considered as the fundamental stages of every DT process. They are: (1) acquisition of data about the problem, (2) idea generation, and (3) testing the ideas (Brown, 2008; Liedtka, 2015). In the other models with more stages, some stages are split up into sub-stages (Beverland et al., 2015; Liedtka, 2015; Brown, 2008). The IDEO process model with five stages (Shapira et al., 2017) and the D.School model with six stages (Da Silva et al., 2020) are frequently used in practice. The processes are usually iterative making it possible to go back and forth between individual stages (Brown, 2008). In the following, we focus on the three-stage conceptualization (Liedtka, 2015) as the least common denominator and relate the stages to the principles and applicable methods, also referring to the three types of design tools according to Seidel and Fixson (2013).

3.2.1 Data gathering

The starting point of DT is the observation and analysis of the situation with a focus on the user (Brown, 2008). It is about understanding the identity of the user and their hidden needs (Furue and Washida, 2017; Nagaraj et al., 2020). A key in this is building empathy, which means recognizing and understanding of other persons' sensations, emotions, thoughts, motives, and personality traits (Brown, 2008; Carlgren et al., 2016b; Glen et al., 2015). This immersion allows to recognize new possibilities and opportunities (Liedtka, 2020). The information is used to create so-called personas, which depict the typical user and their characteristics, actions, and needs (Welsh and Dehler, 2013). DT strives for a holistic view also called gestalt view (Figure 2), a comprehensive understanding of the problem that includes environmental factors, relationships, background, trends, and user needs (Holloway, 2009). The notion of exploring the whole situation by dividing it into parts and analyzing connections and patterns stems from system thinking (Buchanan, 2019). With this knowledge, designers are able to derive the latent user needs (Liedtka, 2015).

Some of the “needfinding tools” (Elsbach and Stigliani, 2018; Seidel and Fixson, 2013) used in this stage stem from ethnography (Beckman and Barry, 2007) and seek direct contact with the user, such as participant observation, interviewing, journey mapping or job-to-be-done analysis (Liedtka, 2015). Consumers can also be actively involved in the process through co-creation (Leavy, 2012; Luotola et al., 2017). Rather than observing users individually, crowdsourcing allows to observe groups (Macdonald and Elahee, 2016; Mount et al., 2020). Stephens and Boland (2015) emphasize the importance of aesthetic knowledge and active involvement of the bodily senses for immersion and understanding of a situation.

3.2.2 Idea generation

In essence, ideation is about developing hypotheses of possible solutions to a problem or user needs (Liedtka, 2020). Framing and reframing the previously collected data helps identify patterns and generate ideas (Beckman and Barry, 2007). It involves looking for hidden problems to better understand the composition of the problem and target points for solutions (Carlgren et al., 2016b). Collaboration among multidisciplinary teams can be a critical success factor when developing ideas by including multiple perspectives and complementary knowledge (Brown, 2008; Li, 2002; Seidel and Fixson, 2013). To generate ideas, teams first use divergent thinking to develop an extensive range of ideas and, second, convergent thinking to analyze and narrow down these ideas (Carlgren et al., 2016b). New solutions emerge from the development of diverse ideas from large stakeholder groups, which are used in the confluence of analysis and intuition to develop new solutions (Nagaraj et al., 2020). Abductive reasoning is key to connecting intuitive and analytical thinking (Martin, 2010) by asking “what if?” (Liedtka, 2015) and “what might be?” (Martin, 2010). As a result of diverse ideas and abductive reasoning, curiosity and openness within DT teams are increased, and ingrained behavioral patterns are broken (Liedtka, 2015).

Typical idea generating tools are associated with brainstorming and visualization (Seidel and Fixson, 2013). Brainstorming requires high team reflexivity by regular discussions about the ideas within the team (Seidel and Fixson, 2013). For brainwriting, concepts are first written down individually and then discussed in a group, whereas speedstorming is inspired by speed dating (Thompson and Schonthal, 2020). Visualization allows presenting the ideas and detecting unnoticed possibilities by making the ideas tangible through illustration by diagrams, drawings, Post-it notes or scribbles on whiteboards (Carlgren et al., 2016b). Mind mapping can connect the mass of information and get a common understanding within the team (Liedtka, 2015). Narratives and storytelling (Liedtka, 2015) or sociodrama (Wyman et al., 2012) are further tools in this stage.

3.2.3 Testing

In the testing stage, the previously developed and selected ideas and assumptions are subsequently converted into prototypes and tested with experiments iteratively by repeating them several times with additional new adjustments until the ideal solution is found (Carlgren et al., 2016b). A principle within this phase is learning in action (Liedtka and Kaplan, 2019). In contrast to prototyping in engineering where a technically advanced test model is created, prototyping represents ideas by creating a product with minimal input to obtain feedback for the concept (Glen et al., 2015). The generate-test cycles allow for quick feedback and new knowledge that is directly reintroduced into new combinations to obtain the best fit between user-need, environment, and product characteristics (Nagaraj et al., 2020). Admitting mistakes offers valuable insights in this regard (Carlgren et al., 2016b).

Idea-testing tools are intended to assess the solutions on three dimensions: the level of desirability for users, the technical feasibility, and the business viability (Elsbach and Stigliani, 2018). Success on these three dimensions provides the opportunity for innovation. Rapid prototyping makes ideas tangible by building small models with little effort and minimal cost (Brown, 2008). In this regard, there are connections to pragmatism by observing which solution works and using the reflections to test new possibilities (Dalsgaard, 2014). A popular tool to test the formulated assumptions from the previous phase with external stakeholders is field experiments (Liedtka, 2015; Micheli et al., 2019).

3.3 Context factors

3.3.1 Application context

The application areas of DT are predominantly in problem-solving and innovation (Dell'Era et al., 2020). The three stages of data acquisition, ideation, and testing allow DT to be used for problems that are particularly difficult to define (Liedtka, 2015). In addition, the application fields can be characterized by uncertainty and ambiguity (Elsbach and Stigliani, 2018; Luotola et al., 2017). Therefore, DT is also used within crises to find innovative solutions (Cankurtaran and Beverland, 2020). Nevertheless, the problem does not necessarily have to be challenging (Nakata and Hwang, 2020). Within research and development, DT can be used in both phases (Magistretti et al., 2021b). Nakata (2020) suggests that DT can be used as a part of product development to be applied in the front-end of the process (Brand et al., 2021), and in the back-end to apply another approach like Stage Gate, which, in contrast, is based on analytical decision-making and milestone planning. DT can also be used to transform an organizational culture by integrating its principles and values, such as user focus, norms of collaboration, risk-taking, ambiguity, experimentation, learning from failure, and design-led strategic thinking (Elsbach and Stigliani, 2018). Case studies demonstrate that DT is applicable in many ways, regardless of the industry, such as in topics like promoting sustainability (Buhl et al., 2019; Geissdoerfer et al., 2016; Shapira et al., 2017), financial services (Vetterli et al., 2016) or in the digital context (Przybilla et al., 2020; Shafiee et al., 2021). At the management level, it can support strategy development (Cagnin, 2018; Knight et al., 2020; Ben Mahmoud-Jouini et al., 2016). In areas such as the commodity industry, which usually has little innovation potential, DT can be applied to develop better service offerings for users (Rau et al., 2017).

3.3.2 Organizational factors

For the implementation of DT, non-designers need to learn the design skills and principles of dealing with ambiguity and uncertainty, developing a holistic view, and collaboration in teams among divergent thinking and convergent thinking (Dym et al., 2006). Therefore, participants must train to overcome cognitive obstacles when conducting the process (Butler and Roberto, 2018). It can take some time for the innovative capabilities of DT to be fully utilized (Ben Mahmoud-Jouini et al., 2019). For the implementation in existing structures, scholars recommended using a facilitator who teaches the design methods, moderates through the stages, and when necessary, mediates between the design process and barriers within the organization (Daniel, 2016; Hölzle and Rhinow, 2019; Starostka et al., 2021; Wrigley et al., 2018). At the organizational level, conflicts can arise when DT interferes with existing processes and structures or does not go hand in hand with the organizational culture (Carlgren et al., 2016a; Coco et al., 2020). Therefore, the organization should have a strategic vision and clear goals known among employees and linked to the DT process (Dunne, 2018; Wrigley et al., 2020). In addition, it is necessary to define precisely the guidelines and responsibilities previously (Carlgren et al., 2016a; Wrigley et al., 2020). When introducing DT into existing projects defined by strict metric goal achievement and efficiency, conflicts may arise between the design process and previous project work (Dijksterhuis and Silvius, 2017; Hölzle and Rhinow, 2019; Nakata, 2020). Nevertheless, scholars suggest not to apply DT as a rigid process but rather as an agile method that provides a variety of tools for different situations and can be applied in different ways (Chen and Venkatesh, 2013). In order to use the full capacity of DT, the majority of employees can be authorized to participate in opportunity finding and should be familiarized with the DT process and tools (Appleyard et al., 2020; Liedtka and Kaplan, 2019).

3.4 Outcomes

3.4.1 Organizational Level

The use of DT can improve the performance of firms (Suci et al., 2021). Regarding the development of products, due to the user-focus, products can be developed that have high utility and meet the needs of users (Chen et al., 2018). In doing so, radical innovations can be encouraged (Radnejad et al., 2020; Tiberius et al., 2021). Furthermore, DT can be used to develop the ability of company's brand ambidexterity (Zheng, 2018), helps with the strategy development (Liedtka and Kaplan, 2019; Holloway, 2009), and strengthens the organizational culture (Snyder et al., 2018), and the formation of dynamic capabilities (Magistretti et al., 2021a). Beverland et al. (2015) showed how DT supports brand ambidexterity, the ability to pursue two different strategic paths, consistency and relevance, simultaneously. This is about exploiting the existing resources available to a company, which can lead to incremental innovations. On the other hand, exploring new possibilities through new knowledge and experimentation can lead to radical innovations (Beverland et al., 2015; Zheng, 2018). Moreover, recent studies indicate the elements of DT can be categorized as dynamic capabilities for innovation, by sensing opportunities, seizing them, and transforming or reconfiguring the resources (Kurtmollaiev et al., 2018; Liedtka, 2020; Magistretti et al., 2021a). At the team level, DT promotes a team's overall performance (Nakata and Hwang, 2020; Suci et al., 2021). In addition, DT team outcomes demonstrate high levels of creativity (Lee et al., 2019; Meinel et al., 2020).

3.4.2 Individual level

On an individual level, the DT process enhances participants' creativity (Lee et al., 2019; Kim, 2020; Cummings and Yur-Austin, 2021; Sándorová et al., 2020). In addition, cognitive biases are released, and inertia is broken, allowing for new directions to be taken (Liedtka, 2015; Nagaraj et al., 2020). It also results in higher levels of psychological empowerment (Roth et al., 2020) and self-confidence (Rao et al., 2021; Liedtka, 2020). Moreover, DT can be helpful to build entrepreneurial skills (Lynch et al., 2021; Sarooghi et al., 2019).

4. Discussion

4.1 Research framework

The systematic literature review provides an organized outline of the growing number of publications on the still evolving DT field. Based on the current literature, we propose a framework that includes the process stages, principles and tools, and that also reflects the context factors and outcomes (Figure 3). Furthermore, the framework can be used to identify research gaps.

4.1.1 Context factors

Several organizational factors can foster the DT process. On the one hand, the necessary equipment has to be provided in physical form with rooms and resources required for the application of DT (Wrigley et al., 2020). On the other hand, DT should not collide with existing processes and align with the organizational culture (Carlgren et al., 2016a). A decisive factor for the conditions that need to be created is how DT is implemented and on which levels it is applied (Starostka et al., 2021). In the case of DT's application as a method for a specific project, the organizational conditions need to be created to prevent DT from colliding with existing guidelines (Hölzle and Rhinow, 2019). Adopting DT as a mindset in an organization affects the strategic vision and organizational culture in addition to structural conditions (Wrigley et al., 2020). The principles need to be in line with the organizational culture. For example, in organizations that focus mainly on efficiency, the culture would have to be adapted to the principles of DT or an independent environment for the design mindset must be created (Dunne, 2018). In addition, elements such as empathy, user focus, creativity, willingness to experiment, openness to failure can enrich the organizational culture and encourage employees to actively seek innovation together (Coco et al., 2020; Elsbach and Stigliani, 2018; Kolko, 2015). The type of leadership style plays a vital role in the implementation of processes. Here, the need for research has not yet been met. An impetus for research on leadership styles is offered by Nakata (2020). She recommends a leadership committed to coaching for DT, which does not restrict employees but encourages them to be creative and supports them. Until now, the factors for the composition of teams and the individual factors have been largely disregarded. Diversity and multidisciplinarity within design teams are essential components for defining problems and generating ideas (Seidel and Fixson, 2013). It is assumed that too large teams will have adverse effects (Thompson and Schonthal, 2020). Here, studies could provide insights into how design teams should be composed.

In contrast to the organizational perspective, research is stall scarce on which characteristics individual participants must have before the DT process, which attributes potentially interfere with design principles, and how this affects the outcome. Participants are expected to build empathy, present their ideas in physical or rhetorical form, and work together in teams (Micheli et al., 2019). Thereby, the question could be explored whether people need specific capabilities to participate in the process and how individual characteristics that are not compatible with DT's mindset can be better integrated.

4.1.2 Design thinking process

The center of the framework describes the DT process. The analysis showed the different interpretations of the individual process steps, which have a creative process for problem-solving in common. Differences are more in the structure and wording than in the actual content. For the framework, a three-stage categorization of the process steps according to Liedtka (2015) was chosen, consisting of the stages of data gathering about user needs, idea generation, and testing. However, the choice of three higher-level stages is not meant to exclude models with multiple sub-steps. Subsequently, the frequently quoted elements of Carlgren et al. (2016b) user focus, problem framing, visualization, experimentation and diversity were classified to the three stages as fundamental principles. The categorization of management tools in this article was adopted from Seidel and Fixson (2013) and Elsbach and Stigliani (2018). The search for a definition indicates three different perspectives on DT in research. This framework combines all three views: (1) process, in a general three-step process structure; (2) mindset, i.e. the principles that form the basis for DT mindset; and (3) tools, which are used to carry out the individual steps. Extensive research exists in determining the characteristics and attributes of DT (Carlgren et al., 2016b; Elsbach and Stigliani, 2018; Liedtka, 2015; Micheli et al., 2019). To better determine the selection of design tools, research is needed on the individual effects of the tools. The possibilities of combinations and connections of methods and tools have also not yet been considered in studies. Here, critical insights for practitioners can emerge. In addition, research could be conducted to answer which tools are most effective in specific situations or industries. Early studies on emerging tools should proceed, including using new technologies, such as augmented and virtual reality (Earle and Leyva-de la Hiz, 2020) or netnography, a digital derivative of ethnography for application in social media (Ashman et al., 2021).

The focus of research on the impact of DT has increased, showing benefits at both the organizational and individual level. In the early years, research often discussed the use and impact of DT through case studies describing its implementation in practice (Beverland et al., 2015; Clune and Lockrey, 2014; Holloway, 2009; Leavy, 2010; Liedtka, 2014; Vetterli et al., 2016). Case studies are still the most commonly used research method in current DT research (Knight et al., 2020; Magistretti et al., 2021b; Mount et al., 2020; Pham et al., 2022; Wrigley et al., 2020). Nevertheless, there was a lack of studies that empirically measured the impact and benefits of DT to give it a firm place as a tool for innovation management (Micheli et al., 2019). This research gap has been initially addressed in past studies. Recent empirical studies were able to confirm the effects previously assumed merely based on case studies. DT has been shown to significantly impact new product development and innovation (Chen et al., 2018; Nagaraj et al., 2020; Nakata and Hwang, 2020). Moreover, it promotes teams' overall performance (Nakata and Hwang, 2020; Suci et al., 2021). Performing DT leads to increased creativity on the group level (Lee et al., 2019) and individual level (Cummings and Yur-Austin, 2021; Kim, 2020). Kim (2020) also showed that the individual creative potential influences the level of innovation. Other individual-level outcomes include increased motivation and empowerment, which is a mediator for project performance (Roth et al., 2020) and increased levels of self-confidence (Liedtka, 2020; Rao et al., 2021). In addition to solving problems, the findings at the individual level indicate that DT is also a means of empowering employees and developing skills, which leads to a better performance.

DT can be integrated into the organization's strategy (Coco et al., 2020) and it can transform the organizational culture based on the characteristics of each stage in terms of user focus, collaboration, risk-taking, ambiguity, experimentation, learning from failure, and design-led strategic thinking (Elsbach and Stigliani, 2018; Kolko, 2015). Moreover, DT can reduce hierarchies and foster better decision-making (Liedtka, 2015). Combining analytical and intuitive thinking can be challenging in organizations that previously relied only on analytical thinking. When introducing DT, persuasion may also be needed to justify its use and opportunities. For this purpose, Dunne (2018) mentioned harvesting the “low-hanging fruits” first by application in projects where DT shows quick success and offers a way to train the participants. During implementation, it is beneficial to introduce a facilitator who can guide the team through the design process and intervene when necessary (Wrigley et al., 2018). Here, it is important to give the participants the confidence and free space to develop new creative ways. In order to take full advantage of the design capabilities, training and time are needed, but this should not be a deterrent because even inexperienced teams can make progress (Seidel and Fixson, 2013).

4.2 Limitations and future research

As with every research, this work comes with several limitations. The basis of the systematic literature review is the dataset of literature on the topic. Here, the inclusion criteria were limited to articles with “design thinking” in the title. The dataset could be extended by including other keywords, such as specific design tools. However, the additional screening of articles with different keywords would have increased the number of articles and thus probably limited the quality of the analysis. For future research with a more specific focus on one of the dimensions of DT, additional keywords would be recommended.

In addition, in the sector analysis, only articles with a direct thematic reference to a particular sector were included. Another approach could be to evaluate all case studies from the articles and determine the industry distribution. The framework provides a very general overview of the context factors, process, and outcome. Additionally, effects due to mediators and moderators within DT should be observed. In the future, it is necessary to explore the interrelationships more precisely in quantitative terms.

DT continues to offer much scope for future research. In general, the versatility of DT should be preserved in the future (Beckman, 2020; Johansson-Sköldberg et al., 2013). Research has started to explore the application of DT in diverse industries. Whereas DT seems to be applicable in many contexts, it is still unclear if differences exist. Additionally, the process and principles of DT could be transferred to other firm activities, such as entrepreneurial marketing (Eggers et al., 2020), which can also be a specific vehicle to market innovative, design-driven products or services.

In his article, Brown (2008) described the characteristics of the design thinker with the five properties empathy, integrative thinking, optimism, willingness to experiment, and collaboration. New research could explore which additional traits are necessary and what weightings the characteristics have for the outcome. This ties into the call for research on how to compose design teams. The insights can help in planning and assembling DT teams. Case studies are often used in research to investigate how DT is applied in practice. Thus, research usually follows an inductive process. Researching more abductively by asking “what might be” could help develop new potentials for DT.

5. Conclusion

In this paper, the DT literature is analyzed. Based on the dataset, it can be concluded that DT has gained increased relevance in research in recent years. We propose a consolidated definition for DT as an iterative innovation and problem-solving process, which is based on specific principles (such as a focus on user needs, multidisciplinarity, etc.) and uses specific methods (such as creative thinking, visualization, experimentation, etc.).

Our proposed framework provides a holistic overview of the context factors, the process, and the outcomes. DT promotes the development of product innovations and the development of innovative behavior and capabilities at the organizational and individual levels. It requires and promotes a culture that allows participants to be creative and considers failures to be important insights. In addition, at the team level, collaboration should be possible, allowing for different opinions and breaking down cognitive biases.

Several research gaps remain. The effectiveness and efficiency of the DT process and its distinct methods should further be tested in empirical studies. Regarding the organizational factors, an urgent need exists for research on leadership styles useful for DT. Research could provide valuable insights related to the composition of teams in terms of team size and degree of interdisciplinarity. Besides, no research was found on individual characteristics that have an influence on the implementation of DT. Although great strides are being made toward DT becoming a respected process in the management literature, there is still a need for future research to unleash its full potential.

Figures

Figure 1

Review process

Figure 2

Attributes of design thinking

Figure 3

Design thinking framework

Table 1

Definitions of design thinking

Author(s)	Definition
Beckman and Barry (2007, p. 25)	A generic innovation process, grounded in models of how people learn
Brown (2008, p. 2)	A discipline that uses the designer's sensibility and methods to match people's needs with what is technologically feasible and what a viable business strategy can convert into customer value and market opportunity
Martin (2010, p. 38)	A balanced dynamic interplay between analytical thinking and intuitive thinking
Seidel and Fixson (2013, p. 19)	The application of design methods by multidisciplinary teams to a broad range of innovation challenges
Glen et al. (2014, p. 653)	An iterative, exploratory process involving visualizing, experimenting, creating, and prototyping of models, and gathering feedback
Beverland et al. (2015, p. 593)	A creative and strategic process characterized by the following hallmarks: abductive reasoning, iterative thinking and experimentation, holistic perspective, and human-centeredness
Kolko (2015, p. 4)	A set of principles collectively known as design thinking—empathy with users, a discipline of prototyping, and tolerance for failure chief among them—is a tool for creating simple, intuitive and pleasurable interactions and developing a responsive, flexible organizational culture
Liedtka (2015, p. 927)	A hypothesis-driven process, that is problem and solution focused, best suited to decision context in which uncertainty and ambiguity are high, composed of empathy, abduction, cocreation and collaboration, visualization and prototyping, and iteration
Shapira et al. (2017, p. 286)	A process and attitude that harnesses creative problem-solving by focusing on the discovery of root causes and needs, collaborating across disciplines, cultivating optimism, and experimenting with solutions in order to learn and adapt more quickly
Elsbach and Stigliani (2018, p. 2274)	Design thinking comprises an approach to problem-solving that uses tools traditionally utilized by designers of commercial products, processes, and environments

Source(s): Authors' own creation

Table 2

Overview of different design thinking process structures

Authors	Process stages
Beckman and Barry (2007)	Observe and Notice		Frame and Reframe	Imagine and Design	Make and experiment
Beverland et al. (2015)	Destabilization		Define and Develop		Transformation
Brown (2008)	Inspiration			Ideation	Implementation
Glen et al. (2015)	Problem finding	Observation	Visualization/sensemaking	Ideation	Prototype and testing	Viability testing
Da Silva et al. (2020)	Understand	Observe	Define	Ideate	Prototype	Test
Shapira et al. (2017)	Discovery		Interpretation	Ideation	Experimentation	Evolution
Liedtka (2015)	Data gathering about user needs			Idea generation	Testing

Source(s): Authors' own creation