2.1 Open Innovation

Open Innovation (OI) was introduced by Chesbrough (Reference Chesbrough2003) and describes the opening of a company’s innovation process to its environment to allow the purposeful exchange of knowledge (Chesbrough, Vanhaverbeke & West Reference Chesbrough, Vanhaverbeke and West2006; Dittrich & Duysters Reference Dittrich and Duysters2007; Gianiodis, Ellis & Secchi Reference Gianiodis, Ellis and Secchi2010; Braun Reference Braun, Braun, Eppinger, Vladova and Adelhelm2012; Chesbrough & Bogers Reference Chesbrough, Bogers, Chesbrough, Vanhaverbeke and West2014). Internal and external knowledge are considered equal in value and usability (Chesbrough Reference Chesbrough2003). The basis for this knowledge exchange is the purposeful collaboration with external partners (West & Gallagher Reference West and Gallagher2006), such as suppliers, customers, research institutes, companies from other industries or even competitors (Huizingh Reference Huizingh2010). In this regard, OI is not a completely new phenomenon, since there are several examples of early precursor forms of OI, which date back long before 2003 (Trott & Hartmann Reference Trott and Hartmann2009; Huizingh Reference Huizingh2010). Thus, we consider OI to be a new methodical framework to structure existing approaches and methods of collaborative innovation, such as the Lead-User approach (von Hippel Reference von Hippel1986) or distributed product development, e.g., Hameri & Nihtilä (Reference Hameri and Nihtilä1997). OI also supports the purposeful enhancement of this by adding new approaches, methods, and tools. In addition to the involvement of individuals and specific companies, OI also fosters the collaboration with unspecific groups of external partners (‘crowds’) from all over the world by using web technologies. In contrast to traditional market research, external partners participate actively and creatively in the innovation process (Hilgers & Piller Reference Hilgers and Piller2009). The ongoing academic dispute between Chesbrough’s company-centric definition of ‘Open Innovation’ and the individual-centric ‘Open User Innovation’ (von Hippel Reference von Hippel2005) is excluded at this point. Interested readers may refer to Chesbrough, Vanhaverbeke & West (Reference Chesbrough, Vanhaverbeke and West2014).

Opening up an innovation process in terms of OI is not a binary decision. A company needs to carefully analyze and select suitable process phases or projects for OI. The suitability for OI depends on different influencing and context factors, e.g., phase of product lifecycle (PLC), competitive market situation, need for concealment, internal processes. To stress the situation-dependent differing levels of openness, Dahlander & Gann (Reference Dahlander and Gann2010, 699) introduced the term ‘permeability’. Considering the flow of knowledge (Gassmann & Enkel Reference Gassmann and Enkel2004), three types of OI can be distinguished, which are consistent with a differentiation based on the location of the innovation (Chesbrough & Crowther Reference Chesbrough and Crowther2006; Chesbrough & Bogers Reference Chesbrough, Bogers, Chesbrough, Vanhaverbeke and West2014):

1. (1) outside-in/inbound OI: transferring external knowledge into the company, strengthening the internal innovation process, and enabling internal innovations,

2. (2) inside-out/outbound OI: giving internal knowledge to the company’s environment to allow external innovations and exploit new markets, and

3. (3) coupled OI: a combination of the above two types.

In addition to this, other categories of differentiation can be also found, such as pecuniary/non-pecuniary (Dahlander & Gann Reference Dahlander and Gann2010) and the role of the company (Gianiodis et al. Reference Gianiodis, Ellis and Secchi2010).

2.2 Situative Open Innovation

In the context of this research, SOI is a methodology for planning OI projects (Guertler et al. Reference Guertler, Holle, Guber and Lindemann2014a ). The goal of SOI is to support teams from industry and academia with a methodology that can either be applied for autonomously planning OI projects or to allow a better evaluation of external service offers. The strategic decision for or against OI as well as the execution of the OI project itself are excluded. SOI consists of five phases, which are grouped into two areas, as illustrated in Figure 2. The outer ring represents the rough planning of the OI project, addressing the question: Who to involve, in which way and why? As illustrated by the arrows, SOI allows iterations, e.g., when new information is available or boundary conditions change, or when matching OI partners and OI methods. To ensure purposeful iterations, stage gates assess and control the planning progress (Cooper Reference Cooper2001). The inner area represents the detailed planning, including aspects such as acquisition of OI partners or specific starts and durations of OI methods.

Figure 2. Situative Open Innovation (Guertler et al. Reference Guertler, Holle, Guber and Lindemann2014a ).

The first phase of SOI 1 is the analysis of the characteristics of the OI project and of the company’s internal and external context factors, such as organizational aspects or dynamics of competitors. Their assessment is the focus of this publication. Based on SOI 1, the SOI 2 phase identifies existing stakeholders as well as new potential OI partners. Potential OI partners are assessed, ranked and selected with regard to their relevance to the OI project. Based on the OI situation and selected OI partners, suitable OI methods are derived in SOI 3 and adapted to the specific project situation. SOI 4 focuses on defining the performance assessment, controlling and risk management of the OI project. SOI 5 comprises the detailed planning of the rough project structure defined in the previous phases, e.g., the acquisition of OI partners or the specific start date of an OI method.

2.3 Situations in innovation management and product development

The basic element of SOI is the analysis of a company’s ‘situation’, i.e., context factors that influence and constrain the OI project. As already stated by Birkhofer, Jänsch & Kloberdanz (Reference Birkhofer, Jänsch and Kloberdanz2005, 9f) and their ‘10 Commandments’ for product design, it is essential to ‘meet the design situation’. Different authors state the relevance of the situation-specific selection and adaption procedures for product development and innovation management in general (Birkhofer et al. Reference Birkhofer, Kloberdanz, Berger and Sauer2002; Gericke et al. Reference Gericke, Meißner and Paetzold2013; Lindemann Reference Lindemann2009, 29; Ponn Reference Ponn2007, 43). However, besides its dynamic character, the term ‘situation’ itself is complex and a general definition is missing (Ponn Reference Ponn2007, 43), which makes the literature review challenging. In addition, while some sources explicitly name criteria and applicable values for assessing a situation (e.g., ‘form of governance’ (Bevis & Cole Reference Bevis and Cole2010, 6)), others only mention influencing factors of the innovation performance, which can be used to derive situation criteria (e.g., ‘centrality of R&D’ (Gassmann, Enkel & Chesbrough Reference Gassmann, Enkel and Chesbrough2010, 213)). Other sources represent case studies, describing specific OI projects, and can also be used to derive situation criteria, e.g., von Hippel (Reference von Hippel1988). As a ‘mixed type’, some authors mention situation-based criteria, but no information about their assessment (e.g., ‘innovativeness’ (Hauschildt & Salomo Reference Hauschildt and Salomo2005), or ‘company culture’ and ‘hierarchies’ (Ponn Reference Ponn2007, 55)). A partial exception is presented by Gericke et al. (Reference Gericke, Meißner and Paetzold2013). They analyze different definitions of the term ‘context’ and present a large collection of different context influencing factors, which might be relevant for the adaption of design methods. However, they neither provide detailed information about those influencing factors nor direct references to primary sources. Besides a lack of an OI focus, there is no analysis of the relevance of the factors in the context of adapting methods. Although the consideration of an industry perspective is stated as being crucial for a successful long-term application, it is missing from the literature so far.

Therefore, in the following, we will analyze different approaches for assessing situations in product development and OI. Based on this, we derive a definition of OI situations. At this point, we will focus on a general description of situation analysis, while single criteria and their references are presented in Section 4.

3.1 Criteria categories for describing an OI situation

The following section presents the structure of the evaluated situation criteria lists, grouped into the four categories derived from the literature as described above. The lists themselves can be found in the appendix A. The structure of the lists aims at supporting easy use by other research groups. Besides the criterion’s name, the tables also contain a criterion ID for a clear identification, a short description of the criterion, potential values based on the literature and evaluation, evaluation remarks, i.e., comments from the workshops and resultant activities, relevant references, and application cases within SOI.

Figure 4 illustrates the basic structure of the criteria lists. Criteria highlighted in dark gray and indicated by an ‘M’ in the application case (AC) column are part of the OI method selection tool. Light gray highlighting indicates criteria which were tested for the DMM situation method mapping, but showed a too weak correlation (for details, see Figure 6). Criteria for reflexive use are marked with an ‘R’. ‘N/A’ indicates criteria that were evaluated as being not applicable and were ‘canceled’ (omitted)/removed from the criteria set.

4.1 Industrial evaluation of usability and measurability

To evaluate the criteria’s usability in terms of comprehensibility and measurability, we evaluated them with three industry partners. The evaluation was conducted in three half-day workshops with small and medium enterprises (SMEs) that were not experienced in OI (in the focus of SOI); the SMEs are in the machine and plant engineering sectors: (1) a supplier of mechanical automotive parts, (2) a manufacturer of production machines and related services, and (3) a manufacturer of building technology products. Each workshop team consisted of three to four employees (specialists and managers) from different departments. In the case of the first company (1), the participants were: the director of corporate development, the manager of the relevant specialized department, an R&D engineer from that same department, and a R&D engineer from a central innovation department. For the second company (2), the OI team consisted of the manager and an engineer from the advance development department, an R&D engineer from the relevant specialized regarding specialty department and an expert from the purchasing department. The OI team from the third company (3) included two managers from the specialized department with differing areas of responsibilities and an expert from that same department, who had previously also worked in sales for a couple of years.

The teams had to fill in the questionnaire lists autonomously prior to the workshop and indicate problematic criteria. This allowed a comparison of expected criterion values with values assessed by the teams from the companies. The evaluation in general focused on practical significance, comprehensibility, and measurability of criteria. Based on this, criteria were modified, concretized, added or removed, and criterion values defined if possible. A cross-case comparison led to the main conclusions. The few cases of different opinions were discussed in a meeting with all participants involved in the situation analysis.

The underlying questions were:

• ∙ Which criteria are useful from a company’s point of view?

• ∙ Which criteria need to be improved in terms of comprehensibility and distinctiveness?

• ∙ Which criteria necessitate major effort in terms of data acquisition, or might be problematic due to other reasons, such as data privacy?

• ∙ Which criteria should be added to the existing list?

4.1.1 Criterion-specific feedback

In the following, we present the criteria-specific results from the evaluation, i.e., which criteria were modified due to which reasons (indicated by an ‘M’ in the criteria list), which were removed/canceled (‘C’), and which were added (‘N’ for new). Unchanged criteria are marked by a ‘U’.

Table 1 shows criteria that were modified as a result of the industry partners’ feedback as well as the underlying reasons.

Table 1. Modified situation criteria.

Criteria which were canceled/removed from the criterion list are depicted in Table 2.

Table 2. Canceled/removed situation criteria.

New criteria added during the workshops are listed in Table 3. The final tables of criteria are depicted in the appendix, Figures 7–10.

Table 3. New, added situation criteria.

4.2 Matrix-based identification of relevant criteria for selecting OI methods

To achieve our overall goal of successfully planning OI projects and our research goal G1 in particular, it is necessary to identify those criteria which are especially relevant for OI. By using a matrix-based approach, we identified criteria that are relevant for selecting suitable OI methods. We derived 27 situation criteria. The full process in the context of developing an OI method selection tool is described in Guertler et al. (Reference Guertler, von Saucken, Tesch, Damerau, Lindemann, Huizingh, Torkkeli, Conn and Bitran2015). In the following, we focus on deriving the OI-relevant criteria.

By means of a workshop-based academic discussion, we pre-filtered the OI criteria list by criteria which provide only background information and do not support the selection of an OI method, such as ‘name of company’, ‘year of founding’, or ‘compulsory certifications’. The resulting 52 situation criteria were subsequently mapped onto OI method criteria by using a Domain Mapping Matrix (DMM) (Danilovic & Browning Reference Danilovic and Browning2004). First, the mapping was conducted on a criteria level to identify relevant dependencies. In a second step, the identified dependencies were further detailed by mapping particular criteria specifications, as illustrated in Figure 5.

Figure 5. Two-step DMM mapping of OI situation and OI method criteria.

The criteria-level mapping was conducted independently by five members of our academic research team, using a four-step scale with zero (no link), one (weak link), two (medium), and three (strong link). Subsequently, the single DMMs were added up, resulting in mapping values ranging from zero to 15. To reduce the influence of subjectivity and singular perceptions, we analyzed a minimum mapping value. Assuming at least a medium-rated mapping value for each assessor, we derived ten as an initial minimum value. To validate this value, we analyzed all mappings with a value between eight and eleven with a detailed mapping on a criteria specification level, as shown on the right side in Figure 5. This revealed that a specification-level mapping is possible for mapping values of nine, but not below. Accordingly, we consider nine to be the minimum mapping value.

Figure 6 shows the pertinent criteria-level DMM. Matrix cells with a mapping value equal or higher than nine are highlighted in gray. Situation criteria are considered relevant for selecting OI methods if the respective matrix rows contain at least one mapping value equal to or higher than nine. An analysis of the excluded situation criteria revealed that (a) they are criteria which only indirectly affect the selection of OI methods, such as (1.6.1) ‘Innovativeness of employees’ and (4.5.2) ‘Strategic location of OI project in the company’, or (b) they address projects in the past, but not the current one, such as (3.3.2) ‘Type of external partners’ and (3.3.4) ‘Type of cooperation’.

Figure 6. DMM mapping of OI situation and OI method criteria (gray: relevant criteria and cells).

Other criteria were excluded as well, although they contained mapping values above the mapping limit. The main reasons were that they (a) are too fine-granular for the selection process, such as (4.1.3) ‘Secondary OI project goal’, or are quite similar to other criteria, such as (3.5.2) to (3.5.1), (4.2.2) to (4.2.1), (4.2.5) to (4.1.2), and (4.4.5) to (2.7.4).

The number of relevant mapping cells are not evenly distributed over the situation criteria. The majority contains only a low number, a minority contains a large number of mappings, which are especially criteria characterizing the OI project.

4.3 Key learnings and guidelines for developing a situation analysis

Summarizing the results from the industrial evaluation workshop and the DMM mapping, we derived the following key learnings, which contribute to our research goal G2. They provide guidance and support to research teams when developing a situation analysis:

• ∙ Is the assessment effort manageable? Optimally, there are no more than 30 criteria, and/or an analysis duration of one day maximum. Too many criteria can demotivate the industry team and lead to a boycott of the situation analysis.

• ∙ Who is able to assess the criteria? The industry partners stated that it could be very demotivating to receive a large number of criteria questions which they are not able to answer themselves, but need to identify colleagues who can. Therefore, it is helpful to state a potential department of the company that is likely to be able to provide the relevant information. In this case, the industry partner can directly forward the criteria to suitable departments and employees.

• ∙ Do you provide an overview of the analysis process? It is important to give an overview of the entire situation analysis process in the beginning as well as continuous updates about the current position in the process and the remaining steps. This lets the industry partners schedule the analysis sufficiently and maintain an overview of the progress.

• ∙ Do you provide a description of each criterion? Often the criteria names are not comprehensible in themselves and require a short description. We also recommended providing a short indication of why and where the criteria will be relevant in the subsequent steps of the methodology.

• ∙ Do you prevent ambiguity? Especially when working with interdisciplinary teams, terms and expressions can have varying meanings or interpretations, e.g., (2.5) ‘Innovation cycles in industry’ could be interpreted as times between two technologies as well as the time period for developing a new product.

• ∙ Are you aware of the dynamics of criteria properties? Generally, the situation criteria assess only one situation state of one particular project. For a subsequent project or even during the progression of a particular project, some criteria might change. To reduce the risk of misplanning, you should identify the most dynamic criteria as well as define a schedule for checking for changes in the properties.

• ∙ Do you use time-independent criteria? Dynamic changes of criteria properties pose a challenge in the situation analysis. Therefore, it is important to define the criteria themselves in a time-independent manner in order to minimize obsolescence, e.g., instead of assessing the age of a company, it is better to assess the year of founding.

• ∙ Do you use defined property scales for each criterion? To reduce effort and the risk of obtaining unusable data and demotivating the industry partners, distinctive property scales should be used. Often property ranges are sufficient, but without corresponding scales, companies might try to assess the criteria as precisely as possible and spend too much unnecessary time and effort.

• ∙ Do you provide references for Likert scales? Often scale properties such as ‘low’, ‘medium’, and ‘high’ are not absolutely defined. For example, the majority of companies might assess (2.7.4) ‘need for concealment’ as ‘high’, but in one case that could mean publishing product concepts are critical, and in another, it could even include mentioning the company’s name. Consequently, providing a reference example for each scale element is beneficial.

• ∙ Do you clarify the organizational and temporal focus level? Some criteria can be assessed on different levels and for different time periods. To avoid ambiguity, it is important to specify the focus level, e.g., OI experience on the level of the company, the business unit, the department, or the team. Define if you are assessing the present, the last 10 years, or other time periods.

• ∙ Do you consider the specific subjectivity of criteria? Some criteria have a high level of subjectivity such as (3.4.1) ‘Employees’ attitude towards external partners’. In those cases, an assessment by additional persons or substitution by other, more objective criteria might be beneficial.

• ∙ Are you aware of inherent risks in the context of individual-related data? Along with subjectivity and dynamics, individual-related data is critical. It is important to check if it is legal to even assess that data, how to ensure confidentiality, and how to prevent offending persons when seeing the assessments (e.g., innovativeness (1.7.1)).

• ∙ Are you aware of potential strategic risks of the criteria? Companies might refuse to assess criteria due to their high strategic relevance and resultant risks if information, e.g., profits or expenses for R&D, gets into wrong hands. To avoid unassessed criteria, you should check why you need those criteria. It might be possible to obtain the underlying information with other, less critical criteria.