Content uploaded by Angelo Cavallo
Author content
All content in this area was uploaded by Angelo Cavallo on May 17, 2025
Content may be subject to copyright.
Industrial and Corporate Change, 2025, 00, 1–23
DOI: https://doi.org/10.1093/icc/dtaf018
Original Article
Internal organization and absorptive
capacity in open innovation contexts: the
roles of gatekeeping, task specialization, and
delegation
Angelo Cavallo1,*, Massimo G. Colombo1 and Nicolai J. Foss2
1Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Via Raffaele
Lambruschini, 4/B, Milan 20156, Italy. e-mail: angelo.cavallo@polimi.it; e-mail: massimo.colombo@polimi.it
and 2Department of Strategy and Innovation, Copenhagen Business School, Copenhagen, Kilevej 12, 2nd ;
2000 Frederiksberg, Denmark. e-mail: njf.si@cbs.dk
*Main author for correspondence.
Firms increasingly face strategic choices in the innovation process concerning how and to what extent they
should source knowledge from external parties. We offer new insight into the organizational design aspect
of this process. Specically, we build a model informed by organization theory and organizational economics
to examine how rms’ returns from absorbing knowledge vary as a function of their choices concerning the
specialization of gatekeepers and the delegation of decision-making authority. We compare returns under
different organizational congurations and contingencies and derive propositions about how rms should
optimally organize to absorb external knowledge. We provide qualitative evidence of how rms organize their
external knowledge absorption activities from four “real-life” cases that illustrate the propositions of our
model.
JEL classication: L10, M21
1. Introduction
Firms increasingly source external knowledge to support their innovation processes (Chesbrough,
2003; Afuah and Tucci, 2012; Tucci et al., 2016; Zobel et al., 2016; Lu and Chesbrough, 2022).
Choices concerning such sourcing, such as which sources to target, the level or intensity of the
sourcing, and how to govern sourcing arrangements, have become key strategic decisions (e.g.,
Laursen and Salter, 2006; Foss et al., 2011; Gambardella and Panico, 2014; Monteiro, 2015;
Tucci et al., 2016; Zobel et al., 2016; Dahlander et al., 2021; Cavallo and Burgers, 2024; Park
et al., 2024). In a parallel line of inquiry, scholars have examined the strategic management of
absorptive capacity (AC) (e.g., Lane and Lubatkin, 1998; Jansen et al., 2005; Miroshnychenko
et al., 2021; Anand and Mukherjee, 2024), usually dened as the “ability to identify, assimilate,
and exploit knowledge from the environment” (Cohen and Levinthal, 1989: 589). While these
two literature streams have developed in parallel, the phenomena of open innovation and AC
may be strategic complements in the sense that the more rms invest in AC, the more they will
gain from investing in open innovation, and vice versa.
© The Author(s) 2025. Published by Oxford University Press in association with the Industrial and Corporate Change
Association.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://
creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any
medium, provided the original work is properly cited.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
2A. Cavallo et al.
In this study, we investigate this complementarity through an organizational lens. Our key
overall idea is that successful management of the complementarity between AC and open inno-
vation requires an organizational design that is geared to support it. For example, companies
such as Lego and IBM decentralized decision-making authority to user-centered project teams
and redesigned reward systems to build an AC that ts an open innovation model (Hienerth
et al., 2012). More generally, research already suggests that in an open innovation context, inter-
nal organization matters for both potential and realized AC (e.g., Jansen et al., 2005; Foss et al.,
2011; Colombo et al., 2013; Backmann et al., 2015; Monteiro, 2015; Colombo et al., 2021).1
However, the emerging literature on how internal organization can support knowledge absorp-
tion in an open innovation context does not offer a precise theoretical framework that predicts
which (internal) organizational design will be chosen to maximize the returns from knowl-
edge absorption, given such contingencies as overall knowledge quality (Monteiro, 2015), the
dynamism of the environment (see Volberda et al., 2010). Instead, the extant literature offers a
“one size ts all” approach based on substantial delegation of decision-making authority over
knowledge absorption to specialized gatekeepers. However, the knowledge that rms seek to
absorb in an open innovation context may differ depending on the characteristics of their envi-
ronments, such as the quality of external knowledge—a fact that should be reected in rms’
internal organizations. For instance, state-of-the-art scientic knowledge held by research scien-
tists that, despite being entirely explicit, is highly complex differs in nature and characteristics
from informal, more tacit product- or process-related knowledge held by customers or equipment
producers (Volberda et al., 2010; Schilke, 2014; Dosi et al., 2022). Knowledge developed in an
emerging technical eld may be of lower average quality and more limited average value than
more proven knowledge developed in more established elds, even if some particular pieces of
knowledge in an emerging eld may be of high value. In turn, these various knowledge environ-
ments may have different implications for internal organization choices, and these implications
may be conditional to the varying knowledge-absorbing abilities of rms’ employees.
The purpose of this study is to add to our understanding of such internal organization choices,
which hold important implications for the understanding of AC and open innovation and, per
implication, economic growth (Nambisan et al., 2018). Building on key ideas in the information-
processing branch of the organization design literature (Marschak and Radner, 1972; Galbraith,
1974; Tushman, 1978; Sah and Stiglitz, 1985, 1986, 1991; Radner, 1993; Dessein, 2002; Des-
sein and Santos, 2006), we examine how the combination of organizational design elements
affect rms’ returns from absorbing external knowledge.2 We specically consider two important
design elements: 1) the specialization of the gatekeeping task and 2) the delegation of decision-
making authority over knowledge absorption to those who perform gatekeeping tasks. These
elements (reecting choices related to organizational structure and decision systems, respectively)
broadly correspond to those identied in the empirical literature on organizational design and
AC. However, in contrast to that stream of literature, we explicitly distinguish between delega-
tion and specialization and consider different combinations of these variables in organizational
congurations. Moreover, our model allows for comparisons of returns not only under different
organizational congurations but also under different contingencies. The key contingencies we
examine are (i) the friendliness of the external knowledge environment, dened as an environ-
ment where external knowledge generally is of high quality (Volberda et al., 2010), and (ii) the
varying abilities to screen external knowledge of individual organizational members (Knudsen
and Levinthal, 2007).
Our model allows us to derive novel and non-intuitive predictions regarding how rms should
organize to absorb external knowledge. We show that the extent of the specialization of the
1The distinction between potential and realized AC refers to the distinction between rms’ to successfully acquire
and assimilate knowledge and their ability to transform and deploy the externally acquired knowledge for use in
production and innovation (Zahra and George, 2002).
2One relevant aspect of organizational design is getting the incentives right in the external organization, that
is, between partners who collaborate in open innovation projects (Gambardella and Panico, 2014). If incentives are
misaligned, less knowledge may be produced that can be absorbed by the parties. However, in this study, we are only
concerned with the design of the rm’s internal organization.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 3
knowledge absorption task in “gatekeepers” and the delegation of authority over external knowl-
edge absorption decisions to lower levels of the corporate hierarchy are indeed inuenced by the
friendliness of the external knowledge environment and the distribution of knowledge-screening
abilities across a rm’s personnel (and their combination). To demonstrate the practical relevance
of the propositions derived from our theoretical model, we provide qualitative evidence from four
real cases that illustrate how rms operating in different knowledge environments and with per-
sonnel equipped with different knowledge-screening abilities organize their external knowledge
absorption activities.
2. Background: organizational design and AC
2.1. The AC literature
Since its introduction by Cohen and Levinthal (1989, 1990), AC has become a key construct
in several elds, including innovation studies, international business, and strategic management.
Over nearly 30 years of research, AC has been treated as an antecedent of other macro vari-
ables, an outcome, and a moderator or mediator (Volberda et al., 2010). The construct itself,
dened initially as the “ability to recognize the value of new information, assimilate it, and
apply it to commercial ends” (Cohen and Levinthal, 1990: 128), has been unpacked in terms
of the “organizational routines and processes by which rms acquire, assimilate, transforms and
exploit knowledge to produce a dynamic organizational capability” (Zahra and George, 2002),
as reected in the distinction between potential AC and realized AC.
Volberda et al. (2010) review 1213 papers in the AC literature and identify a number of
research gaps. They argue that research on AC should “be explicit about what kind of knowledge
is being absorbed” (Volberda et al., 2010, p. 943); “address the varying nature of knowledge”
(p. 943); explain the “impact of individuals on the AC process” (p. 944) and the “origin of
organization-level AC” (p. 945); address “how AC existing on different levels of analysis (indi-
vidual, organizational, dyadic, etc.) are related” (p. 945); and “systematically explore how formal
organization inuences the level, formation, and dynamic nature of AC and the retrieval of prior
knowledge” (p. 946). Thus, rms source different kinds of external knowledge from different
external knowledge sources, and not only may implications for returns differ depending on what
knowledge rms’ source from which sources, but different types of knowledge and different
sources may also require different organizational designs for successful absorption of that knowl-
edge. However, the extant research does not offer systematic insights into how managers can best
structure organizations, deploy mechanisms of organizational control, and train employees to
optimize the absorption of external knowledge given various contingencies, such as the quality
of the external knowledge that the rm seeks to absorb.3
2.2. Internal organization and AC
In their breakthrough paper, Cohen and Levinthal explicitly sought to ground rm-level AC
in individual-level considerations (1990: 131). However, they also pointed out that an orga-
nization’s AC is not merely the aggregation of individuals’ absorptive abilities. Specically, AC
depends on transfers of knowledge across and within subunits in the organization, so that an
understanding of AC requires an understanding of “the structure of communication between the
external environment and the organization, as well as among the subunits of the organization,
and also on the character and distribution of expertise within the organization” (1990: 131–2).4
3While AC theory per se pays little attention to organizational members, related theorizing highlights members
and, in particular, the roles assumed by individuals in organizations, such as boundary spanning and gatekeeping (Allen,
1977; Tushman and Katz, 1980; Katz et al., 1995; Soh and Roberts, 2005; see Barzilai-Nahon, 2009, for a review).
However, while such work provides insights into acquiring and assimilating knowledge, it speaks less directly to the
other key processes of AC, especially the transformation and exploitation of this knowledge. An understanding of these
processes also requires that attention be paid to individuals and their roles.
4Cohen and Levinthal (1990) also introduced a distinction between “outward-looking AC” and “inward-looking
AC.” The former refers to the actual acquisition of externally held knowledge, and the latter refers to the internal transfer
and exploitation of the acquired knowledge (see also Lenox and King, 2004). In the context of knowledge acquisition,
they explicitly referred to Allen’s (1977) work on gatekeepers (i.e., individuals, often with a managerial role, who
manage the informational interface between the organization and its environment).
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
4A. Cavallo et al.
These authors related knowledge acquisition directly to issues of centralization and the delegation
of decision-making authority. For example, they pointed out that centralization of gatekeeping
functions may not be efcient when external knowledge changes quickly and erratically: “Under
such circumstances, it is best for the organization to expose a fairly broad range of prospective
‘receptors’ to the environment. Such an organization would exhibit the organic structure of Burns
and Stalker (1961: 6).” In sum, they explicitly highlighted individuals, organizational practices
and design, and the nature of the “knowledge environment” in which the organization nds itself.
A small stream of empirical literature has picked up on these suggestions and empirically
linked organizational design variables to AC (Van den Bosch et al., 1999; Jansen et al., 2005;
Foss et al., 2011; Hienerth et al., 2011; Keinz et al., 2012; Asmussen et al., 2013; Colombo et al.,
2013, 2021; Monteiro, 2015). However, few studies explicitly consider AC in an open innovation
context (e.g., Jansen et al., 2005, does not, while Foss et al., 2011, and Colombo et al., 2021,
do).
In addition, extant studies mainly examine the marginal effects of single organizational design
variables on AC and do not consider the effects of alternative congurations of such variables.
For instance, Colombo et al. (2021) nd that under open innovation conditions, the nature of
knowledge (they distinguish between “scientic” and “practical” knowledge) shapes the main
effect relation between the rm’s R&D intensity (a measure of the strategic importance of R&D)
and the rm’s delegation of decision rights to R&D personnel, such that scientic knowledge
turns the negative main effect relation positive, while practical knowledge makes it stronger.
However, they only consider a single organization design variable (i.e., delegation).
Finally, key contingencies, such as the quality of the external knowledge that rms may seek to
absorb and the screening abilities of gatekeepers, are generally not considered, perhaps because
they are difcult to measure.5 However, intuitively, such parameters matter in the choice of
organizational design in the context of knowledge absorption. For example, if a rm is in an envi-
ronment where knowledge is predominantly of bad quality and such knowledge can be harmful
to rm performance (e.g., the relevant knowledge is associated with a string of false positives),
and the rm suspects that the knowledge is not reliable, it would seem paramount to carefully
screen knowledge (Knudsen and Levinthal, 2007). This may call for some employees to specialize
in gatekeeping, as such specialization improves their ability to screen external knowledge. Should
these employees also have the right to make decisions regarding using insourced knowledge in
production? On the one hand, they specialize in screening knowledge. On the other hand, a more
hierarchical organization of external knowledge screening may be required when the relevant
knowledge is predominantly of bad quality (Sah and Stiglitz, 1985, 1991).
Such issues are not easy to resolve by relying purely on verbal logic. Hence, in the following,
we build a simple formal model that can address and resolve important aspects of designing the
internal organization for knowledge absorption in an open innovation context.
3. Organizational designs for external knowledge absorption
3.1. Basic assumptions
To reduce complexity, we make some simplifying assumptions. We consider a rm that is engaged
in only two operating tasks: 1) insourcing external knowledge (i.e., the gatekeeping task involv-
ing knowledge acquisition and assimilation), and 2) deploying the insourced knowledge in the
context of innovation and production (i.e., the knowledge transformation and exploitation task,
which involves the combination of the insourced external knowledge with internally generated
knowledge, and the exploitation of the combined knowledge for innovation and production). In
doing so, we assume the rm’s stock of internally generated knowledge is given and freely accessi-
ble to the two employees. Hence, while a rm’s knowledge base does inuence the attractiveness
of absorbing external knowledge, as originally emphasized by Cohen and Levinthal (1989), it
5There are a few exceptions. For example, while examining the relation between selective attention and knowledge
absorption from the environment in multinational corporations, Monteiro (2015) highlights dimensions of “knowledge
provenness” and “knowledge dissonance.” Colombo et al. (2021) show that in an open innovation context, the associ-
ation between R&D intensity and the level of delegation of decision-making authority over innovation decisions varies
depending on the type of knowledge (scientic or practical) rms aim to absorb.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 5
does not inuence the organizational design of knowledge absorption. The insourcing and deploy-
ment tasks are complements in the sense that, in the absence of either task, the value of the rm’s
output is zero. This assumption captures Zahra and George’s (2002) idea that a potential AC
is a precondition for innovation but does not guarantee the transformation and exploitation of
the insourced knowledge. However, the two tasks are substitutes in terms of employee time, as
employees allocate a xed “time budget” across external knowledge insourcing and knowledge
deployment.
The rm has two hierarchical layers. At the bottom of the rm’s hierarchy, two employees
(A and B) are engaged in operating tasks (i.e., external knowledge insourcing and knowledge
deployment) and report to a manager. The manager performs strategic tasks and may double-
check employees’ decisions related to the operating tasks. Strategic activities are only performed
by the manager, who is not involved in the operating tasks. The employees receive a xed salary.6
The manager’s use of time to double-check employees’ decisions has an opportunity cost, as
the manager devotes less time to strategic activities (Harris and Raviv, 2002). In accordance
with the information-processing stream of the organizational design literature (Marschak and
Radner, 1972; Galbraith, 1974; Sah and Stiglitz, 1985), we assume that the two employees and
the manager make decisions that maximize the value of the rm’s output given their information
sets (i.e., we assume away incentive conicts).
The value of the rm’s output (Y) is a function of the value of the external knowledge that
is absorbed (K) and the time employees spend on knowledge deployment for innovation and
production (T), such that Y=TxK. We normalize the sum of the time each employee spends on
operating tasks to 1. K depends on the amount and quality of the external knowledge employees
insource. If an employee devotes all his time to external knowledge insourcing, he can process
N units of external knowledge. Therefore, the quantity of external knowledge insourced by the
rm depends on how much time employees devote to this task.
3.1.1. The friendliness of the external knowledge environment
While external knowledge can be dimensionalized in different ways (e.g., Winter, 1987; Rogers,
2003), for simplicity we assume there are two types of external knowledge: high-quality knowl-
edge and low-quality knowledge. Each unit of high-quality external knowledge employees
insource generates a return for the rm equal to z1> 0 per unit of time they devote to knowledge
deployment for innovation and production. Conversely, insourcing a unit of low-quality external
knowledge results in a negative return for the rm equal to -z2 per unit of time in knowledge
deployment. The share of high-quality external knowledge relative to the total amount of avail-
able external knowledge is . Given the values of parameters z1 and z2, reects the “friendliness”
of the external knowledge environment.7 With no loss of generality, we henceforth assume that
z1=z2=1.
The probability that a given piece of external knowledge is screened and judged to be bene-
cial for the rm by the employees handling the knowledge insourcing task (i.e., the gatekeepers)
depends on the quality of that knowledge (while keeping constant the screening abilities of
employees, see below). The probability of accepting high-quality external knowledge is lower
than unity (but greater than ½), while the probability of accepting low-quality external knowl-
edge is greater than 0 (but lower than ½). This means that there is always a non-zero probability
that: 1) high-quality external knowledge is discarded even though it should have been in-sourced
by the rm (i.e., a non-zero probability of a type I error; see Sah and Stiglitz, 1986), and 2) low-
quality external knowledge is in-sourced even though it should have been discarded (i.e., a type
6In our model, the salary of employees is a sunk cost for the rm and, therefore, is not considered further.
7For example, a dynamic environment where technical change is very rapid and university scientists are the key
source of external knowledge (like nano-materials) is very unfriendly. While some pieces of external knowledge may
be very valuable to rms, most of them are not. Conversely, in a technologically mature environment where customers
and producers of materials and equipment are the main sources of knowledge for rms, most knowledge is of good
quality, even if the value created by absorbing each individual piece of knowledge is limited. The friendliness of the
external knowledge environment is related to the β parameter in Cohen and Levinthal (1989: 572). They assume
that an increasing β reects an increasing marginal effect of R&D on AC but a declining effect on the level of R&D,
indicating that when external knowledge becomes more difcult to absorb, the rm will absorb less of it (for a given
level of R&D).
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
6A. Cavallo et al.
II error). We assume that the expected returns from absorbing external knowledge are positive
despite the negative returns from absorbing low-quality external knowledge.
3.1.2. The external knowledge screening abilities of employees
The two employees, A and B, may differ in their abilities to screen external knowledge, that is,
insourcing good quality knowledge and discarding bad quality one. Without loss of generality, we
assume that employee A has better knowledge screening abilities than employee B if there are dif-
ferences. For example, employee A may have better human capital (i.e., better education, greater
work experience) or more social capital, which provides better access to high-quality “soft” infor-
mation on external knowledge. Employees’ screening abilities are reected in parameter k (we
will discuss this in greater detail later).
3.2. Internal organization
We consider two key elements of a rm’s organizational design that relate to organizational struc-
ture and decision systems: (i) the specialization of the operating tasks (i.e., the gatekeeping and
knowledge deployment tasks), and (ii) the delegation of decision-making authority over exter-
nal knowledge insourcing. We acknowledge that organizational structure and decision systems
encompass much more than these two elements. However, these two elements have been explic-
itly identied in the empirical AC literature as antecedents of AC (e.g., Foss et al., 2011; Colombo
et al., 2013).
3.2.1. Task specialization
When tasks are specialized, employee A specializes in the gatekeeping task and devotes its time
entirely to external knowledge insourcing. Employee B specializes in knowledge deployment by
combining the external knowledge employee A insources with the rm’s existing knowledge and
exploiting this combination for innovation and production. Suppose decision-making authority
over knowledge absorption is delegated to employees. After screening external knowledge. In that
case, A insources it and transmits it to employee B, who uses it for innovation and production. If
decision-making authority is centralized in the manager’s hands, employee A needs the manager’s
authorization to transmit knowledge to employee B.
We assume that there is no learning-by-doing from specializing the gatekeeping task. How-
ever, task specialization offers two potential advantages. First, if employee A has better screening
abilities than employee B, the likelihood of both type I and type II errors decreases with task spe-
cialization. Let r1 and r2 be the probabilities that the specialized gatekeeper accepts high-quality
and low-quality external knowledge, respectively. p1 and p2 are the corresponding probabilities
when tasks are not specialized. We assume r1=kp1 and r2=1
kp2, with 1 < k, so that r1>p1
and r2<p2.8 Second, knowledge-related economies of scale arise from task specialization (Lei-
jonhufvud, 1986; Argote and Epple, 1990). As the specialized gatekeeper devotes all his or her
time to external knowledge insourcing, he or she processes N units of external knowledge. When
the gatekeeping task is not specialized, we assume that the two employees devote half of their
time to this task and use the remaining half of their time for knowledge deployment. Accordingly,
each of them can process and eventually use N/2 units of external knowledge for innovation and
production. Hence, while keeping the screening ability of employees constant and assuming that
there is no knowledge sharing between employees, the total amount of external knowledge used
in production is half the amount used when tasks are specialized.
Despite the advantages of task specialization illustrated above, a potential drawback is that
employee A must pass on the insourced external knowledge to employee B, who then uses it for
knowledge transformation and exploitation. Knowledge transmission from A to B is subject to
losses, distortions (Keren and Levhari, 1979, 1983, 1989), and delays (Radner, 1993; Bolton and
8Let t1 and t2 be the probabilities that employee B accepts external knowledge of good and bad quality, respec-
tively. It results: t1= ((2 k)/k)r1≤r1 and t2= (2k−1)r2≥r2. If k were equal to 1, there would be no gain from task
specialization. As mentioned earlier, we assume that the expected returns of external knowledge absorption are positive
regardless of whether tasks are specialized. That is, βr1—(1—β)r2> 0 and βp1—(1—β) p2>0.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 7
Dewatripont, 1994; Van Zandt, 1998). In our model, we consider these inefciencies through the
parameter < 1, which reects the deterioration of the external knowledge used for innovation
and production that occurs when tasks are specialized. Factor corresponds to the “efciency
factor” Zahra and George (2002) dened as the ratio of realized to potential AC. In other words,
the communication of insourced knowledge within the rm, and the associated losses, distortions,
and delays are sources of the divergence of realized AC from potential AC, which hinders the
effective use of high-quality external knowledge.9 If there is no deterioration of the insourced
external knowledge (i.e., = 1), realized AC equals potential AC.
3.2.2. Delegation of decision-making authority
Decision-making authority concerning external knowledge absorption may be delegated to
employees or centralized in the hands of the manager. In the former case, the employees han-
dling the gatekeeping task (i.e., employee A if tasks are specialized; both employees if they are
not) decide autonomously whether the knowledge they have insourced is to be used in combina-
tion with existing knowledge for innovation and production and the manager rubberstamps their
decisions. If decision-making authority is centralized, the employees performing the gatekeeping
task submit proposals to the manager concerning insourcing external knowledge and deploying
it for innovation and production. The manager double-checks the quality of the knowledge. For
the sake of simplicity, we assume that the manager’s screening abilities are equal to the average
screening abilities of the two employees. External knowledge is only used following the man-
ager’s approval. The advantage is that in order for low-quality external knowledge to be deployed
for innovation and production, it must be mistakenly judged as high-quality twice (i.e., by the
employee submitting the proposal and by the manager approving it). Hence, the centralization
of decision-making authority over external knowledge absorption reduces the likelihood of type
II errors.
However, such centralization has two drawbacks. First, it increases the likelihood of type I
errors. In fact, under centralization, high-quality external knowledge is absorbed by the rm
only if its quality is judged as high twice (cf. Sah and Stiglitz, 1986). Second, the examination of
employees’ proposals is costly in terms of the manager’s time. We assume that examining a given
proposal takes one unit of the manager’s time, which is diverted from his or her other duties (e.g.,
making valuable strategic decisions) at an opportunity cost equal to .10
3.3. Comparing congurations of internal organization: results
In Figure 1, we synthesize the congurations that emerge from combinations of the two organi-
zational design elements discussed in the previous section. We are interested here in examining
how the choice of the organizational design conguration depends on: (1) the “friendliness” of
the external knowledge environment, captured by the parameter ; (2) the advantage in screening
external knowledge of employee A (i.e., the specialized gatekeeper in case tasks are specialized),
captured by the parameter k; and (3) their interplay. Firms prefer conguration i over congura-
tion j if the returns associated with the adoption of conguration i, as reected in the value of the
rm’s output (Y), are greater than those associated with conguration j (i, j= 1,….,4, ij). Hence,
we compare the value of the rm’s output when adopting these different organizational design
congurations for external knowledge absorption. For the sake of synthesis, we rst compare the
circumstances under which rms prefer specialization over non-specialization of the knowledge
insourcing task under the assumption that the rm has a decentralized decision system in which
the manager rubberstamps employees’ decisions (i.e., we compare congurations 1 and 2; see
Figure 1). Second, we examine the circumstances under which the delegation of decision-making
authority over external knowledge absorption to employees is preferable to the centralization of
9Note that in our model, when tasks are not specialized, α equals 1, as knowledge is deployed by the same
employees who in-sourced it.
10 For the sake of simplicity, we assume that manager’s double-checking of an employee’s decision about external
knowledge absorption does not lead to any deterioration of that knowledge.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
8A. Cavallo et al.
Figure 1. Organizational design congurations for absorbing external knowledge. Yk: Returns to the rm of
adopting organizational design conguration k.
that decision-making authority in the manager’s hands. Again for the sake of synthesis, this com-
parison is made under the assumption that tasks are specialized (i.e., we compare congurations
1 and 3).11
3.3.1 Comparing specialization and non-specialization of tasks
Let Y1 be the value of the rm’s output when it adopts an organizational design conguration
with specialization of the gatekeeping and knowledge deployment tasks and decision-making
authority over knowledge absorption delegated to employees. Y2arethe returns to the rm in the
absence of task specialization. The expressions of Y1 and Y2 are as follows:
Y1=[r1−(1 − )r2]and
Y2= 21
2[p1−(1 − )p2]1
2.
Letting 1−2 be the difference between Y1 and Y2 gives:
1−2 0
a=1
2r1
k−(1 − )kr2
r1−(1 − )r2
. (1)
Expression (1) suggests that specializing the gatekeeping task leads to an increase in the value
of the rm’s output if the parameter , which reects the efciency of internal communication
(i.e., the efciency with which the external knowledge insourced by the gatekeeper is transmitted
to the other employee who is engaged in knowledge deployment), is above a given threshold,
a.
Therefore, the lower
a, the greater the likelihood that rms nd it preferable to specialize the
gatekeeping task.
Let us rst consider the impact of the screening advantage of a specialized gatekeeper (com-
pared to the situation where the knowledge insourcing task is not specialized), as reected by
parameter k. We nd that:
−
a
k= − 1
2r1
1
k2+ (1 − )r2
[r1− (1 − )r2]2< 0.
As might be expected, the greater the ability of the specialized gatekeeper to assess the quality
of external knowledge relative to the corresponding ability of the other employee, the lower the
threshold
a and the greater the likelihood that the rm will opt to specialize the gatekeeping task.
11 The results obtained from comparing congurations 3 and 4 and congurations 2 and 4 are similar to those
illustrated in the text. The reason is that in our model, the specialization of tasks and the delegation of decision-making
authority over knowledge absorption have independent effects on rms’ returns. These results are available from the
authors upon request.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 9
It is more interesting to investigate how the threshold avaries with parameter , which cap-
tures the friendliness of the external knowledge environment, as reected in the likelihood that
external knowledge is of high quality. It is easy to show that aincreases with , making the
specialization of the gatekeeping task less preferable:
=r1r2
2[r1− (1 − )r2]2k−1
k> 0. (2)
Moreover, this gives:
2
k =r1r2
2[r1− (1 − )r2]2(1 + 1
k2) > 0. (3)
Given these results, we offer the following propositions:
Proposition 1: The specialization of the gatekeeping task becomes more preferable to the rm
the less friendly the external knowledge environment is.
Proposition 2: The negative association between the friendliness of the external knowledge envi-
ronment and the specialization of the gatekeeping task is stronger the greater the screening ability
of a specialized gatekeeper, compared to the other employees.
According to Proposition 1, in an environment where external knowledge is predominantly
of high quality, there is no point in specializing the gatekeeping task. In this type of knowl-
edge environment, the screening advantage of a specialized gatekeeper has negligible effects on
rms’ returns from knowledge absorption because of the high quality of (most of) the insourced
knowledge. Moreover, the unavoidable communication inefciencies if external knowledge is not
insourced by the same individuals who use it for knowledge deployment are highly detrimental in
this context. These inefciencies are eliminated if tasks are not specialized. The costs of inefcient
internal communication decrease and the gains from the knowledge screening ability of a special-
ized gatekeeper increase with a decrease in the proportion of high-quality external knowledge.
Hence, all else equal, the less friendly the knowledge environment, the more the balance tilts in
favor of task specialization.
Proposition 2 adds that this effect is magnied if specialized gatekeepers enjoy a substantial
advantage over other employees in external knowledge screening. The reason is that the screening
advantage the rm obtains from specialization of the gatekeeping task is particularly valuable if
the likelihood of insourcing high-quality knowledge is low.
In sum, in a friendly external knowledge environment, we expect to observe less special-
ization of the gatekeeping task regardless of the screening ability of a specialized gatekeeper.
Conversely, task specialization will be observed more frequently in an unfriendly knowledge envi-
ronment, especially if specialized gatekeepers have considerably better screening abilities than
other employees. We illustrate these relations in Figure 2.
3.3.2 Comparing delegation and centralization of decision-making authority
With the centralization of decision-making authority (and task specialization), the returns to the
rm from external knowledge absorption, as reected in the value of the rm’s output (Y3), are
given by the difference between the benets the rm reaps from external knowledge absorption
when the manager double-checks the gatekeeper’s proposals, and the opportunity cost of the time
the manager devotes to double-checking employees’ decisions:
Y3=r2
1
1
k−(1 − )r2
2k−[r1+(1 − )r2].
Let 1−3 indicate the difference between the value of the rm’s output with either a decentralized
decision system (Y1) or a centralized system (Y3) (and task specialization). Then we obtain:
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
10 A. Cavallo et al.
Figure 2. Probability of observing specialization of the gatekeeping task. yS=Probability of observing specialization
of the gatekeeping task over external knowledge absorption; k=Advantage of gatekeepers over production
workers in screening external knowledge; 𝛽=Friendliness of the external knowledge environment.
1−3 =r1+(1 − )r2+r11 − r1
1
k−(1 − )r21 − r2k .
Hence:
1−3 0
d= −r11 − r1
1
k−(1 − )r21 − r2k
r1+ (1 − )r2
. (4)
In line with Harris and Raviv (2002), expression (4) posits that delegation of decision-making
authority over external knowledge absorption is preferable when the opportunity costs of a
manager’s time are greater than a threshold
d. It is easy to show that:
−
d
k= −1r121
k2+ (1 − )r22
[r1+ (1 − )r2]< 0 (5)
and
−
d
= −r1r2(2 − r1
1
k−r2k)
[r1+ (1 − )r2]2< 0. (6)
Expressions (5) and (6) indicate that the threshold level of the opportunity cost of the manager’s
time above which the delegation of decision-making authority is preferable to centralization
decreases the higher the screening abilities of a specialized gatekeeper compared to the other
employees and the more friendly the knowledge environment are. Indeed, in our model, if spe-
cialized gatekeepers enjoy an advantage over other employees in screening external knowledge,
they also enjoy a screening advantage over the manager. Then, in line with Hayek’s (1948) con-
tention that information and decision-making authority need to be co-located, the greater this
screening advantage, the more preferable delegating decision-making authority over external
knowledge absorption to the specialized gatekeeper would be. Moreover, as highlighted by Sah
and Stiglitz (1986), double-checking decisions has limited value for the rm in a more friendly
external knowledge environment, as external knowledge is likely to be of good quality. This, in
turn, makes the delegation of decision-making authority preferable to centralization.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 11
Figure 3. Probability of observing delegation of decision authority over external knowledge absorption.
yD=Probability of observing delegation of decision-making authority to the gatekeepers over external knowledge
absorption; k =Advantage of gatekeepers over production workers in screening external knowledge;
𝛽=Friendliness of the external knowledge environment.
It also results:
2−
d
k = −r1r2r1
1
k2−r2
[r1+ (1 − )r2]2< 0. (7)
This leads to the following Proposition:
Proposition 3: The positive association between the friendliness of the knowledge environ-
ment and the delegation of decision-making authority over external knowledge absorption to a
specialized gatekeeper is stronger the greater the specialized gatekeeper’s knowledge-screening
ability compared to the other employees.
Proposition 3 indicates that the screening advantage of specialized gatekeepers relative to other
employees (and managers) reinforces the effect of the friendliness of the external knowledge
environment in favoring the delegation of decision-making authority over external knowledge
absorption. This also means that in a very unfriendly knowledge environment, decision-making
authority is likely to be centralized in the manager’s hands regardless of the screening ability of
a specialized gatekeeper. Conversely, in a friendly knowledge environment, we would expect to
more frequently observe the delegation of the gatekeeping task, and for this effect to increase
with the knowledge-screening ability of the specialized gatekeepers. We illustrate these relations
in Figure 3.
4. Putting the model to use: illustrative evidence
In the following, we bring our theoretical model closer to empirical reality by arguing that one can
use it to predict the internal organizational design that rms adopt to absorb external knowledge
in real-world situations. The previous discussion highlights the quality of the external knowledge
that rms can absorb from external sources like universities, customers, suppliers, and other rms
(i.e., the friendliness of the knowledge environment) and the varying abilities of their employees
to assess that knowledge, resulting in a knowledge screening advantage of some employees (i.e.,
the gatekeepers) over others, as the relevant parameters inuencing the organization of external
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
12 A. Cavallo et al.
Figure 4. Conguration of the organization of knowledge absorption activities.
knowledge absorption. For example, research indicates that different knowledge sources (e.g.,
universities and suppliers) produce knowledge with different characteristics (e.g., tacit or codi-
ed; abstract or practical; more or less focused) and that the quality of knowledge varies over
an industry’s life cycle (e.g., Winter, 1987; Foss and Pedersen, 2002; Rogers, 2003; Haas and
Hansen, 2007). Similarly, a rm’s ability to assess the knowledge generated by an external source
(i.e., the ability to discard low-quality knowledge while detecting and insourcing high-quality
knowledge) depends on the human and social capital of its employees12 and its organizational
practices (e.g., job rotation, job design). In our model, these aspects are captured by parameters
and k and likely differ across rms. Our model predicts that the way rms organize external
knowledge absorption should differ accordingly.
Ideally, one would subject the theoretical propositions developed in the rst part of this arti-
cle to empirical tests. Unfortunately, the requisite data are not available. Therefore, our task is
less ambitious—we use the model to understand better how four rms organize their external
knowledge absorption activities regarding the degree of specialization of the external knowledge
screening task and the delegation of decision-making authority regarding the absorption of exter-
nal knowledge downward the corporate hierarchy. We selected two rms operating in a friendly
environment (i.e., Anna and Bella) and the other two operating in an unfriendly environment (i.e.,
Siemens and JetBlue). We provide illustrative evidence of the predictions of our model based on
these four real-world cases. Figure 4 illustrates predictions and synthesizes how the characteristics
of the organizational congurations of the four rms considered here—that is, the combinations
of the specialization of the gatekeeping task and the delegation of decision-making authority
over external knowledge absorption—vary across the four rms according to the friendliness of
their knowledge environments and the knowledge-screening advantages of some employees over
others.
4.1. Data collection
Our collection of data took place from February 2021 to April 2022. We used various data
sources, which we listed in Table 1.
12 For instance, if rms intend to absorb knowledge from universities or public research centers, employees who
spent part of their professional lives in these organizations are ideally placed to perform this task.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 13
Table 1. Data sources
Data type Quantity Data source
Semi-standardized
interviews
(face-to-face)
27 Anna: 12 interviews with 8 informants, including the CEO, the chair-
man of the board, 3 engineers from the production function, and 3
engineers from the technical-commercial function
Bella: 15 interviews with 9 informants, including the CEO, the inno-
vation manager, the innovation specialist, 4 innovation champions,
and 2 business-area directors
Follow-up interviews
(by phone)
6 2 with the CEO of Anna
2 with the innovation manager of Bella
2 with the innovation specialist of Bella
Internal documents 127 (pages) Presentations, memos, reports (Anna: three documents and one
report; Bella: three documents and one report)
External documents
and sources
153 (pages) Website, news articles, industry reports (with reference to all four
cases)
First, we collected secondary information from rms’ websites, journal articles, industry
reports, and previous studies. Siemens and JetBlue are well-known, large, public rms and are
widely covered by the media. Therefore, we collected data on these rms only from secondary
sources. Specically, our analysis is informed by two recently published business cases regarding
innovation processes at Siemens and JetBlue (Golden et al., 2019; Lakhani et al., 2019).
Secondary sources cover Anna and Bella less well than the other two rms. Therefore, we also
used primary information collected through interviews with several managers and employees
involved in innovation (Eisenhardt, 1989; Siggelkow, 2007; Yin, 2009).13 The interviews lasted
30–140 min (60 min on average) and resulted in 72 pages of transcripts. Each interview was
recorded and transcribed within two days to maintain high-quality data (Gibbert et al., 2008). We
asked the interviewees from Anna and Bella to describe how innovation processes were organized
(e.g., personnel involved, their tasks, and responsibilities), paying special attention to external
knowledge absorption. To reduce misunderstandings, we provided informants with a glossary
containing denitions of key terms derived from the literature. The semi-structured interview
format also allowed interviewees to ask for further explanation at any point during the interviews.
To increase the reliability of our ndings, we complemented the interview data with archival
data from internal documents, such as presentations, reports, memos, and meeting minutes. We
conducted six follow-up telephone interviews to check the validity of our initial ndings (Lincoln
and Guba, 1985).
4.2. Data analysis
Our data analysis aims to provide qualitative evidence illustrating the propositions of our theo-
retical model. For this purpose, we created individual write-ups that triangulated all of our data
(Jick, 1979; Eisenhardt, 1989; Eisenhardt et al., 2016). We then undertook a within-case analy-
sis of each rm before conducting a “cross-case” analysis (Yin, 2009). Specically, in line with
the core logic of comparing opposing organizational design congurations in different environ-
mental settings (see Figure 4), we cross-compared the organization of the external knowledge
absorption processes of the rms operating in friendly environments (i.e., Anna and Bella), and
cross-compared the rms operating in unfriendly environments (i.e., Siemens and JetBlue). More-
over, we replicated the comparison between theoretical propositions and data (Glaser and Strauss,
1967) by examining each case as a standalone observation. This was instrumental in verify-
ing whether the illustrative evidence matched our theoretical propositions. Next, we present the
results of our data analysis.
13 The interviewees requested and were granted anonymity.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
14 A. Cavallo et al.
4.3. The case studies
4.3.1 Anna: the company
Anna is a family-owned and managed small-medium enterprise (headed by the third and fourth
generations of the founder’s family). Founded in the early 1920s, the company began producing
and selling wire cloth and mesh near a major Italian city. By the time of our study, it had evolved
into a specialized provider of metal components for industrial and architectural applications,
operating globally in 60–80 countries with a workforce of over 200 employees. In 2020, the
company reported revenues in the range of EUR 50–70 million.
4.3.2. Anna’s organization of external knowledge absorption activities
Anna operates in a traditional business, where it sells metal components to companies in different
industries. It has three main business areas: (i) lter elements, (ii) conveyor belts, and (iii) products
for the petroleum, petrochemical, and chemical industries. Its competitive advantages mainly
relate to product quality and process efciency. In this respect, it is important for Anna to be
close to the needs of customers, who are a crucial source of external knowledge. The knowledge
provided by customers is fundamental to continuously improving its product offering. Hence,
Anna’s knowledge environment can be considered as friendly.
In each business area, the rm has a U-shaped organizational form composed of a technical-
commercial function and a manufacturing function. The technical-commercial function manages
customer relationships throughout the sales and post-sales processes and oversees business devel-
opment activities. In particular, employees in the technical-commercial function are ideally
positioned to understand customers’ needs and problems, collect customers’ suggestions, and
seize new business opportunities, which often entail incremental improvements in the rm’s prod-
ucts and production processes. In spite of the advantage of technical-commercial employees in
screening external knowledge, the gatekeeping task is not specialized. To assess the feasibility of
incremental innovations, the employees in the technical-commercial function closely collaborate
with the technical employees in the manufacturing function. These latter employees govern all
production processes involved in the realization of the company’s products. When innovation
projects are recognized as valuable and feasible by employees in both functions, the heads of the
two functions commit a budget for their development and assign the projects to mixed develop-
ment teams. This arrangement reects the strong belief of Anna’s top management that effectively
absorbing customers’ knowledge requires close collaboration between the technical-commercial
and manufacturing functions. In the words of Anna’s chief executive ofcer (CEO): “We need
them both [the two functions] to be committed to and responsible for the innovation projects,
as we need to ensure alignment between external demands and internal capabilities.” The CEO
articulated the reason for not specializing the gatekeeping function: “Our industry is stable, and
there has not been any radical revolution around our core business yet. We just need to listen to
our customers about what they need and determine what we can or cannot do.” In other words,
as indicated by Proposition 1, while operating in a friendly knowledge environment, there is no
point for Anna to specialize the gatekeeping task in spite of the knowledge screening advantage
of technical-commercial employees.
The interviews revealed that Anna’s top management prefers to delegate decisions on external
knowledge absorption to teams composed of technical–commercial and manufacturing employ-
ees. The personnel included in these teams possess the right skills for the job. “By working
together, the employees of the two functions have all the needed skills. They talk with customers
every day, while we do not, and they deal with technical issues and suppliers in the produc-
tion plants every day, while we do not” (chairman of the board, Anna). Hence, because of the
friendliness of Anna’s knowledge environment and the screening skills of the employees perform-
ing the gatekeeping task, the probability that the rm will insource and deploy useless (or even
detrimental) external knowledge is very low.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 15
4.3.3. Bella: the company
Bella is a family-owned company operating in a well-established industry. Its size is larger than
Anna. Established in the late 1960s in Italy, the rm initially focused on breeding and selling poul-
try for local markets. Over the decades, it expanded signicantly. At the time of our study, Bella
ranked among the leading poultry producers in Europe, with over 20 business areas, a work-
force of approximately 8000–10 000 employees, and revenues ranging between EUR 1 billion
and EUR 1.5 billion in 2020.
4.3.4. Bella’s organization of external knowledge absorption activities
The most important sources of external knowledge for Bella are materials and equipment sup-
pliers. Their knowledge is very practical and focused on Bella’s innovation processes, so Bella’s
knowledge environment can again be considered friendly.
Throughout its history, Bella engaged in continuous improvements and incremental inno-
vations without a dedicated structure. However, only a short time before our study, its top
management had recognized that continuous improvements were becoming more difcult due
to the company’s large size. In 2019, Bella’s top management decided that it was time to establish
an innovation function. The function was initially composed of an innovation manager and an
innovation specialist. They were responsible for supervising and coordinating several initiatives
to internally promote a culture of innovation, including creating a training program to develop
transversal knowledge on the main innovation trends relevant to the company. They were also
responsible for screening the external knowledge environment, identifying innovation opportu-
nities, and proposing them to the different business areas (e.g., through collaboration with other
rms). However, the company’s top management rapidly realized that a small and specialized
innovation function was ineffective in engaging the business areas in which innovation ultimately
occurred. Therefore, to engage the business areas in external knowledge absorption, management
enlarged the innovation function to include one or more representatives of each business area
whom the area director selected. Thus, 29 “innovation champions” joined the innovation func-
tion as representatives of the 23 business areas. While the innovation champions represent their
respective business areas, they do not have any specic advantage in screening external knowl-
edge compared to other colleagues in their areas. Interviews with the innovation manager and the
innovation specialist indeed revealed that the innovation champions did not possess all the skills
needed to assess the overall feasibility of introducing and implementing a new idea. As stated
by the innovation specialist: “We have innovation champions who have been selected based on
their curiosity and propensity for innovation, but they have diverse backgrounds and compe-
tences. Some of the innovation champions have more commercial backgrounds; others more of
an engineering background. Some of them have experience in different business areas, while oth-
ers have only been in a single business area.” The innovation specialist added: “We know that
they may overlook some important aspects while performing the gatekeeping task just because
they do not have the right experience or knowledge.”
While the innovation manager and the innovation specialist are the “orchestrators” (i.e., they
are responsible for managing the innovation function), the innovation champions are assigned
to select and lter ideas and knowledge relevant to their respective business areas. However, at
Bella, the gatekeeping task is not specialized. Our interviews revealed that the innovation cham-
pions are expected to spend 10–20% of their time on the gatekeeping task, while they use the
rest of their time for other tasks related to their role in the business area. In this regard, an inno-
vation champion stated: “I am an operations manager. My ordinary job—80% of my time—is
to make sure our operations in the Bio-products business area are running smoothly. I dedicate
the remaining 20% to innovation (which includes external knowledge absorption).” The gate-
keepers’ lack of specialization was a deliberate choice by top management, which conforms to
Proposition 1. Innovations in the rm’s industry were mainly incremental, and the most valuable
external knowledge was provided by suppliers and, to a lesser extent, customers. The CEO stated:
“We are keen to pursue innovation, but we know our place. We operate in a traditional industry
where innovation is incremental in processes and products and results from dialogues and dis-
cussions with customers and suppliers. This is why we need a direct link between innovation and
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
16 A. Cavallo et al.
our existing business.” The fact that innovation champions do not have any sizable advantage in
absorbing external knowledge over their colleagues reinforces this effect.
At Bella, members of the innovation function have limited decision-making authority over
the absorption of external knowledge. To implement new ideas, they need the formal approval
of their business-area directors. Moreover, when the new ideas involve more than one business
area—which generates organizational complexity in the subsequent phases of the innovation
process—obtaining the opinion of the “innovation committee” is a mandatory intermediate step
in the knowledge absorption process. The innovation committee, which is composed of the
top management and directors of each business area, meets every second month. Innovation
champions are in charge of presenting new ideas to the innovation committee. If the innovation
committee approves an idea, then one or more business areas step in to handle the execution.
For an innovation project to be executed, there must be one or more “sponsor” business areas
that would commit the necessary budget and take responsibility for the project’s execution and
related business results. For instance, the company collaborated with a startup that, through a
proprietary articial intelligence algorithm, offered a dynamic pricing solution based on products’
expiration dates with the aim of reducing food waste. The innovation committee was involved in
evaluating the relevance of that solution, as it was of interest to different products and business
areas.
4.3.5. Cross-case comparison: organizing external knowledge absorption in a friendly
knowledge environment
As mentioned above, Anna and Bella operate in traditional industries where technological tra-
jectories are consolidated, innovation tends to progress incrementally, customers and suppliers
are the fundamental sources of external knowledge, and the knowledge environment is friendly.
In line with Proposition 1, the gatekeepers’ level of task specialization is low in both cases. At
Anna, the employees in the technical-commercial and production functions are not dedicated
exclusively to external knowledge absorption. Rather, their main tasks are to produce or sell the
rm’s products. Similarly, in Bella, the innovation champions are not specialized gatekeepers.
They spend most of their time running the existing activities in their respective business areas.
Moreover, our interviews revealed similar rationales for this organization of knowledge absorp-
tion. The top management of both companies was aware of the friendliness of the knowledge
environment and employees involved in the existing business were viewed as the most suitable
channel for external knowledge absorption.
The cross-case analysis also shows some differences. The screening abilities of the personnel
handling the gatekeeping tasks in Anna and Bella differed. In accordance with Proposition 2,
the decisions regarding whether to delegate decision-making authority over external knowledge
absorption to those employees also differed. In Anna, the employees in the technical-commercial
and manufacturing functions are highly skilled engineers who have been trained in economic and
managerial aspects and have high knowledge screening abilities. Therefore, the company opted to
delegate decision-making authority over knowledge absorption to employees in these functions.
Conversely, the 29 innovation champions in Bella had more limited screening abilities. Some
innovation champions had an engineering background and were competent in technical issues
but they lacked a full understanding of commercial issues. Others had a managerial-economic
background and more pronounced commercial skills but they lacked the technical skills needed
to assess the technical feasibility of innovative ideas. The lower screening abilities of Bella’s per-
sonnel explained why Bella’s management delegated less decision-making authority to these staff
members compared to Anna.
4.3.6. Siemens: the company
Siemens is a German multinational diversied corporation headquartered in Munich with branch
ofces in several countries. Over the years, the company rapidly grew through organic expansion,
alliances, and acquisitions. At the time of our study, Siemens operated in four main areas: energy,
healthcare, industry, and infrastructure and cities. In 2021, the Siemens Group invested EUR 4.9
billion in R&D, generated revenue of EUR 62.3 billion, reported a net income of EUR 6.7 billion,
and had 303 000 employees worldwide.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 17
4.3.7. Siemens’ organization of external knowledge absorption activities
Siemens operates in high-tech businesses where companies devote substantial resources to R&D
and sources of external knowledge are heterogeneous, including university scientists, innovative
startups, and rms from a variety of sectors. This makes the external knowledge environment
unfriendly.
Siemens has a long tradition of creating, protecting, and commercializing innovations through
(internal) R&D activities. In 2020, it employed 40700 researchers worldwide and led approxi-
mately 23 patents per day (https://new.siemens.com/global/en/company/innovation.html). More
recently, Siemens has championed open innovation. For example, it made its Additive Manu-
facturing (AM) Network and 3D printers available to the global medical community to speed
up the design and production of medical components in response to the COVID-19 pandemic
(Dahlander and Wallin, 2020).
Most internal R&D activities are run at the business area level, while a small number are han-
dled by the Corporate Technology function (CT). Specically, CT is a corporate-level innovation
unit that has more than 7000 employees. Most of them are scientists and engineers with great
external knowledge absorption abilities. In accordance with Propositions 1 and 2, the members
of the CT function operate as specialized gatekeepers who fully focus on exploration activities
and the knowledge absorption process (Lakhani et al., 2019). For instance, CT employees played
an important role in absorbing new knowledge on diverse technologies, such as sensor systems
for extreme environments, electrical grid power storage, and the removal of trace contaminants
from water. The CT function has its own budget (around EUR 6 million in 2019) for scouting ini-
tiatives.14 In particular, the unit is in charge of organizing innovation contests in specic domains
that are of interest to Siemens’ business areas. For example, the unit organized a contest—open
to everyone—on Smart Grid business models, applications, and disruptive technologies, through
which it collected 448 new ideas. A jury of energy experts, including Siemens’s top executives,
ultimately selected the winners, each receiving a EUR 15 000 grant.
While the CT function is responsible for scouting external innovative ideas, decision-making
authority over knowledge absorption is centralized in the top management of the business areas
or in Siemens’ managing board, which is composed of the CEO, the Chief Financial Ofcer, and
the Chief Technology Ofcer. The R&D projects orchestrated by the CT unit are funded directly
by the business areas (approximately 70% of the total budget), centrally funded (approximately
30% of the total budget), or both. For example, in the Smart Grid Contest, the 10 most promising
ideas were funded by Siemens’ Smart Grid business. This organizational design arrangement is
hardly surprising given the unfriendliness of Siemens’ external knowledge environment.
4.3.8. JetBlue: the company
Founded by David Neeleman in 1999 in Long Island City, New York, JetBlue Airways is among
the largest US airlines. At the time of our study, it provided air travel services to 105 destinations
across 22 countries, with most of its operations in the US and the Caribbean. In 2021, JetBlue
generated USD 6037 billion in revenue and had 18 485 employees worldwide.
4.3.9. JetBlue’s organization of external knowledge absorption activities
From its early days, JetBlue leveraged customer service and innovation to distinguish itself from
competing low-cost carriers. For instance, it always used new aircraft, and passengers could
enjoy leather seats and live satellite TV for no additional charge. In 2021, JetBlue introduced
Paisly (JetBlue Annual Report, 2021), a new travel website that leveraged smart technology to
provide customers with individually tailored offers, such as hotel stays and car rentals, based on
their itineraries.
In 2014, JetBlue’s top management realized that the knowledge environment was rapidly
evolving with the emergence of digital and articial intelligence technologies as well as disruptive
innovations in tourism rms’ business models (e.g., the rise of the sharing economy pioneered by
14 The CT unit was also in charge of other open innovation initiatives. For instance, it created the Siemens Tech-
nology Accelerator, which aimed to transform non-core technologies into business opportunities by selling or licensing
them.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
18 A. Cavallo et al.
Uber and Airbnb). With its external knowledge environment becoming increasingly unfriendly,
JetBlue adopted a new strategy to leverage external sources of innovation. According to JetBlue’s
Chairman Joel Petersen, “[i]nnovation is JetBlue’s DNA. And we are still doing a fair amount of
innovation with our technology folks, but we can’t see everything that is going on in the travel
world” (Golden et al., 2019, p. 6). JetBlue created its own corporate venture capital fund (JTV)
as part of this strategy. While JTV was legally and physically separate from the parent company
(i.e., it was located in California) and had operational autonomy (i.e., it had the right to approve
investments up to a certain amount), it had limited strategic autonomy in absorbing external
knowledge. A seasoned JetBlue top manager (i.e., Bonny Simi, Vice President for Talent) ran JTV
from the beginning. Investments (with the exception mentioned above) are monitored and need
the approval of a ve-member Investment Committee, where four of the members are top man-
agers in JetBlue (the Chief Financial Ofcer, the Chief Digital and Technology Ofcer (CDTO),
the General Counsel, and the president of the Travel Products unit). Moreover, JTV’s investment
strategy and operational processes are closely integrated with JetBlue’s planning process.
While JTV employs a small team of specialized investment managers, JetBlue’s top managers
strongly encourage employees at all levels to collaborate with JTV in external knowledge absorp-
tion. As Eash Sundaram, JetBlue’s CDTO, noted: “JTV is a platform and toolkit available for
anyone and everyone in the company to leverage” (Golden et al., 2019, p. 16). Indeed, JTV
employees do not have any specic advantage in screening external knowledge compared to Jet-
Blue employees, as deep knowledge of JetBlue’s operations is fundamental to assessing the value
of this knowledge for the company. In accordance with Proposition 2, JetBlue has limited special-
ization of the gatekeeping task. JTV employees operate as “translators” for JetBlue’s employees,
who play the role of ultimate gatekeepers while devoting most of their time to other tasks within
the rm.
How JetBlue organizes knowledge absorption and the respective roles of JTV’s and JetBlue’s
employees can be best observed in the “innovation sprint” process, a 12-week program dedicated
to scouting external knowledge. Each sprint starts with the JTV team asking JetBlue employees
about their business needs and problems they need to x. The JTV team then selects 100–300
new ventures that might solve some of JetBlue’s problems Then JetBlue’s business-area directors
“pick the six” nalists they would like to meet (Golden et al., 2019, p.17), after which they
select the best new ventures (one or two) with which to work on a proof of concept. In the
end, JetBlue’s business-area directors and employees are responsible for the tasks of insourcing
external knowledge, translating it, and deploying it in the rm’s operations, while JTV’s personnel
play the “facilitator” role.
4.3.10. Cross-case comparison: organizing external knowledge absorption in an unfriendly
knowledge environment
Both Siemens and JetBlue are in unfriendly knowledge environments, albeit for different reasons.
Siemens operates at the forefront of innovation in several high-tech industries. These indus-
tries are characterized by high technological turbulence and rapid technical change, driven by
dynamism in the underlying science base and new technological trajectories (e.g., the energy
transition). JetBlue operates in a traditional industry (i.e., travel services), which has been rejuve-
nated by the emergence of new business models (e.g., in connection with the sharing economy) as
well as new technologies related to articial intelligence and digital transformation (e.g., the y-
ing taxi service by Uber Elevate and other players). In both cases, we observe the centralization of
decision-making authority over external knowledge absorption at the top management level, with
limited decision autonomy granted to the gatekeepers. In Siemens’ case, decisions to introduce
innovations based on external knowledge require top management approval and budget com-
mitment at the corporate or business-area level. Similarly, in the JetBlue case, decision-making
authority over external knowledge absorption is centralized, rst, in the hands of the investment
committee and, second, among business area directors, who have the ultimate say in introducing
innovations in their areas.
Siemens and JetBlue differ in the screening abilities of the personnel performing the gatekeep-
ing task, and their gatekeepers’ level of specialization in external knowledge absorption varies
as well, in accordance with Proposition 2. In Siemens’s case, most of the 7000 employees in
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 19
the CT function are skilled scientists and engineers. As reported by Lakhani et al. (2019: 3),
the CT unit “made contributions to the entire value chain…it conducted innovation projects for
electrical mobility, sustainable urban development, and biotechnology.” Hence, the CT unit’s
personnel possess high knowledge screening abilities. This goes hand-in-hand with a high level
of specialization of the gatekeeping task. Conversely, at JetBlue, the JTV team is composed of a
few people, including three analysts and two investment principals. JTV’s small investment team
specializes in the initial screening of external knowledge. However, knowledge absorption is a
collective effort that involves employees from JetBlue’s different business areas who deal with
external knowledge absorption on a part-time basis. These employees do not have sophisticated
screening abilities, and JTV’s personnel support them in the external search for innovative ideas
and new technologies. As reported by Sundaram (JetBlue’s CDTO), “it takes a while for people
internally to get comfortable with these new technologies” (Golden et al., 2019: 17).
5. Concluding discussion
5.1. Contribution to theory
AC research and open innovation research are closely related research elds with a strong link:
rms need some level of AC to successfully engage in open innovation. This warrants examining
the organization of AC for the purpose of successfully engaging in open innovation. We speci-
cally focus our modeling of the organization of rms’ external knowledge absorption activity in
terms of the specialization of the gatekeeping task and the delegation of decision authority over
this activity.
Although our model is simple, it is still sufciently rich to capture key aspects of how the qual-
ity of external knowledge and the screening abilities of gatekeepers compared to other employees,
inuence the best organizational designs for the purpose of absorbing knowledge from the exter-
nal environment. In addition, it encompasses non-trivial predictions (P1–P3) relating to how the
interplay of these variables leads to non-linearity in rms’ organizational design. We use four
real-life cases to illustrate how the propositions of our model translate in rms’ organization of
the external knowledge absorption process.
5.2. Future research
This study is one of the rst to formally address the organizational design challenge of successfully
engaging in open innovation (see also Gambardella and Panico, 2014). Further progress in this
area will stem from a disciplined dialogue between empirical work and formal theory (cf., Nelson
and Winter, 1982: Chapter 2). In this spirit, we have offered some highly preliminary evidence in
the form of four illustrative case studies. While these cases do not qualify as rigorous evidence,
they provide an “existence check.” In other words, our theoretically derived propositions t at
least some identiable rms. In terms of taking this study further in the empirical dimension, our
propositions can, in principle, be turned into hypotheses and tested, even if the model and the
propositions involve variables that present operationalization and measurement challenges (e.g.,
external knowledge quality and employees’ knowledge screening abilities). We would welcome
empirical research that successfully addresses these challenges and goes beyond the qualitative
supporting evidence we provide here.
With respect to theoretical work, we see several opportunities for future development of the
model. Our model assumes that the stock of rms’ internal knowledge, which inuences their
effectiveness in absorbing external knowledge, is given and freely available to all employees.
An interesting extension involves endogenizing the generation of internal knowledge. One may
assume that there are three tasks instead of two: (i) internal knowledge production, (ii) external
knowledge insourcing, and (iii) knowledge combination and deployment for innovation and pro-
duction. The model would include a strategic choice (how much time the rm’s employees devote
to external knowledge absorption vs. internal knowledge generation), in addition to the organi-
zational design choices (task specialization and delegation of decision authority). If employees’
abilities to generate new knowledge and insource external knowledge are not perfectly aligned,
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
20 A. Cavallo et al.
the organization of internal knowledge generation likely inuences the organization of external
knowledge absorption and vice versa.
Moreover, while extensions of our model in the direction of adding more employees, more
managerial layers, and so on are unlikely to change the basic results of the model (as captured in
P1–P3), the addition of more managerial practices may. For example, one may easily introduce
employee training as a factor inuencing employees’ ability to identify and evaluate external
knowledge. In addition, how rms organize external knowledge absorption may inuence the
screening abilities of employees over time. For example, employees who specialize in the gate-
keeping task may develop over time superior abilities in screening external knowledge, because
of learning-by-doing effects.
A more fundamental challenge relates to integrating loss-of-control (moral hazard) issues with
the model. Agency issues can be brought into the picture by allowing gatekeepers to derive private
benets from knowledge absorption. However, agency issues go beyond this. For example, Cyert
and Kumar (1996) suggest that gatekeepers may have vested interests in particular technologies
owing to, for example, their educational backgrounds. Agency problems may also arise in con-
nection with the internal transfer, transformation, and use of external knowledge, and hence they
clearly deserve further treatment.
Finally, an attractive extension would be to place the model more explicitly in the context of a
dynamic, competitive environment. For example, as dominant designs emerge (Suarez & Utter-
back, 1997), competition typically changes from product to process features, and technological
fervor is supplanted by relative stability and predictability. This inuences a number of the vari-
ables in our model and, hence, what is the most suitable organizational design. Important steps
toward placing the absorption of external knowledge in the context of a dynamic environment
can be found in Baldwin et al. (2006).
Acknowlegments
We acknowledge the nancial support provided by the the PRIN 2022 project “Digitization
and the inclusiveness of entrepreneurial nance”, Prot. 2022Y3AWN5, nanced by the Ital-
ian Ministry of University and Research, and the GRINS project, nanced by PNRR (Piano
Nazionale di Ripresa e Resilienza, Missione 4 (Infrastruttura e ricerca), Componente 2 (Dalla
Ricerca all’Impresa), Investimento 1.3 (Partnership Estese), Tematica 9 (Sostenibilità economica
e nanziaria di sistemi e territori).
References
Afuah, A. and C. L. Tucci (2012), ‘Crowdsourcing as a solution to distant search’, Academy of Management
Review, 37(3), 355–375.
Allen, T. J. (1977), Managing the Flow of Technology, MIT: MA.
Anand, G. and U. K. Mukherjee (2024), ‘Learning from failures: differentiating between slip-ups and knowledge
gaps’, Organization Science, 35(6), 2095–2116.
Argote, L. and D. Epple (1990), ‘Learning curves in manufacturing’, Science, 247(4945), 920–924.
Asmussen, C. G., N. J. Foss and T. Pedersen (2013), ‘Knowledge transfer and accommodation effects in multi-
national corporations: evidence from European subsidiaries’, Journal of Management, 39(6), 1397–1429.
Backmann, J., M. Hoegl and J. L. Cordery (2015), ‘Soaking it up: absorptive capacity in interorganizational new
product development teams’, Journal of Product Innovation Management, 32(6), 861–877.
Baldwin, C., C. Hienerth and E. Von Hippel (2006), ‘How user innovations become commercial products: a
theoretical investigation and case study’, Research Policy, 35(9), 1291–1313.
Barzilai-Nahon, K. (2009), ‘Gatekeeping: a critical review’, Annual Review of Information Science and
Technology, 43(1), 1–79.
Bolton, P. and M. Dewatripont (1994), ‘The rm as a communication network’, The Quarterly Journal of
Economics, 109(4), 809–839.
Cavallo, A. and J. H. Burgers (2024), ‘Building more entrepreneurial organizations through external innovation
contests’, Journal of Product Innovation Management, 42(3), 582–613.
Chesbrough, H. W. (2003), Open Innovation, Harvard Business School Press: Boston
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 21
Cohen, W. M. and D. A. Levinthal (1989), ‘Innovation and learning: the two faces of R & D’, The Economic
Journal, 99(397), 569–596.
Cohen, W. M. and D. A. Levinthal (1990), ‘Absorptive capacity: A new perspective on learning and innovation’,
Administrative Science Quarterly, 35(1), 128–152.
Colombo, M. G., N. J. Foss, J. Lyngsie and C. R. Lamastra (2021), ‘What drives the delegation of innovation
decisions? The roles of rm innovation strategy and the nature of external knowledge’, Research Policy,
50(1), 104134.
Colombo, M. G., E. Piva and C. Rossi-Lamastra (2013), ‘Authorising employees to collaborate with communities
during working hours: when is it valuable for rms?’, Long Range Planning, 46(3), 236–257.
Cyert, R. M. and P. Kumar (1996), ‘Economizing by rms through learning and adaptation’, Journal of Economic
Behavior & Organization, 29(2), 211–231.
Dahlander, L., D. M. Gann and M. W. Wallin (2021), ‘How open is innovation? A retrospective and ideas
forward’, Research Policy, 50(4), 104218.
Dahlander, L. and M. Wallin (2020), ‘Why now is the time for “Open Innovation”’, Harvard Business Review
Boston, MA: Harvard Business Publishing
Dessein, W. (2002), ‘Authority and communication in organizations’, The Review of Economic Studies, 69(4),
811–838.
Dessein, W. and T. Santos (2006), ‘Adaptive organizations’, Journal of Political Economy, 114(5), 956–995.
Dosi, G., M. C. Pereira, A. Roventini and M. E. Virgillito (2022), ‘Technological paradigms, labour creation and
destruction in a multi-sector agent-based model’, Research Policy, 51(10), 104565.
Eisenhardt, K. M. (1989), ‘Building theories from case study research’, Academy of Management Review, 14(4),
532–550.
Eisenhardt, K. M., M. E. Graebner and S. Sonenshein (2016), ‘Grand challenges and inductive methods: rigor
without rigor mortis’, Academy of Management Journal, 59(4), 1113–1123.
Foss, N. J., K. Laursen and T. Pedersen (2011), ‘Linking customer interaction and innovation: the mediating role
of new organizational practices’, Organization Science, 22(4), 980–999.
Foss, N. J. and T. Pedersen (2002), ‘Transferring knowledge in MNCs: the role of sources of subsidiary
knowledge and organizational context’, Journal of International Management, 8(1), 49–67.
Galbraith, J. (1974), ‘Organization design: an information processing view’, Interfaces, 4(3), 28–36.
Gambardella, A. and C. Panico (2014), ‘On the management of open innovation’, Research Policy, 43(5),
903–913.
Gibbert, M., W. Ruigrok and B. Wicki (2008), ‘What passes as a rigorous case study?’, Strategic Management
Journal, 29(13), 1465–1474.
Glaser, B. and A. Strauss (1967), ‘Grounded theory: the discovery of grounded theory’, Sociology the Journal of
the British Sociological Association, 12(1)27–49.
Golden, J. N., A. Sridharan and R. A. Burgelman (2019), JetBlue Technology Ventures: Bringing External
Innovation in House, Stanford Graduate School of Business: Stanford
Haas, M. R. and M. T. Hansen (2007), ‘Different knowledge, different benets: toward a productivity perspective
on knowledge sharing in organizations’, Strategic Management Journal, 28(11), 1133–1153.
Harris, M. and A. Raviv (2002), ‘Organization design’, Management Science, 48(7), 852–865.
Hayek, F. A. (1948), ‘The use of knowledge in society’, In F. A. Hayek (eds), Individualism and Economic Order.
University of Chicago Press: Chicago, pp. 77–91.
Hienerth, C., P. Keinz and C. Lettl (2011), ‘Exploring the nature and implementation process of user-centric
business models’, Long Range Planning, 44(5-6), 344–374.
Jansen, J. J., F. A. Van den Bosch and H. W. Volberda (2005), ‘Managing potential and realized absorptive
capacity: how do organizational antecedents matter?’, Academy of Management Journal, 48(6), 999–1015.
JetBlue Annual Report (2021). Retrieved from: https://blueir.investproductions.com//media/Files/J/Jetblue-IR-
V2/Annual-Reports/jblu019-jetblue-airways-2021-annual-report-web.pdf
Jick, T. D.. (1979), ‘Mixing qualitative and quantitative methods: triangulation in action’, Administrative Science
Quarterly, 24(4), 602–611.
Katz, R., M. Tushman and T. J. Allen. (1995), ‘The inuence of supervisory promotion and network location on
subordinate careers in a dual ladder RD&E setting’, Management Science, 41(5), 848–863.
Keinz, P., C. Hienerth and C. Lettl (2012), ‘Designing the organization for user innovation’, Journal of
Organization Design, 1(3), 20–36.
Keren, M. and D. Levhari (1979), ‘The optimum span of control in a pure hierarchy’, Management Science,
25(11), 1162–1172.
Keren, M. and D. Levhari (1983), ‘The internal organization of the rm and the shape of average costs’, The
Bell Journal of Economics, 14(2), 474–486.
Keren, M. and D. Levhari (1989), ‘Decentralization, aggregation, control loss and costs in a hierarchical model
of the rm’, Journal of Economic Behavior & Organization, 11(2), 213–236.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
22 A. Cavallo et al.
Knudsen, T. and D. A. Levinthal (2007), ‘Two faces of search: alternative generation and alternative evaluation’,
Organization Science, 18(1), 39–54.
Lakhani, K., K. Hutter, S. Pokrywa and J. Fuller (2019), Open Innovation at Siemens, Harvard Business
Publishing: Boston
Lane, P. J. and M. Lubatkin (1998), ‘Relative absorptive capacity and interorganizational learning’, Strategic
Management Journal, 19(5), 461–477.
Laursen, K. and A. Salter (2006), ‘Open for innovation: the role of openness in explaining innovation
performance among UK manufacturing rms’, Strategic Management Journal, 27(2), 131–150.
Leijonhufvud, A. (1986), ‘Capitalism and the factory system’, R. N. Langlois, ed., 1986, Economics as a Process:
Essays in the New Institutional Economics, Cambridge University Press 203–223 : Cambridge
Lenox, M. and A. King (2004), ‘Prospects for developing absorptive capacity through internal information
provision’, Strategic Management Journal, 25(4), 331–345.
Lincoln, Y. S. and E. G. Guba (1985), ‘Establishing trustworthiness’, Naturalistic Inquiry, 289, 289–327.
Lu, Q. and H. Chesbrough (2022), ‘Measuring open innovation practices through topic modelling: Revisiting
their impact on rm nancial performance’, Technovation, 114, 102434.
Marschak, J. and R. Radner (1972), Economic Theory of Teams, Yale University Press: New Haven
Miroshnychenko, I., A. Strobl, K. Matzler and A. De Massis (2021), ‘Absorptive capacity, strategic exibility,
and business model innovation: empirical evidence from Italian SMEs’, Journal of Business Research, 130,
670–682.
Monteiro, L. F. (2015), ‘Selective attention and the initiation of the global knowledge-sourcing process in
multinational corporations’, Journal of International Business Studies, 46(5), 505–527.
Nambisan, S., D. Siegel and M. Kenney (2018), ‘On open innovation, platforms, and entrepreneurship’, Strategic
Entrepreneurship Journal, 12(3), 354–368.
Nelson, R. R. and S. G. Winter (1982), An Evolutionary Theory of Economic Change, Harvard University Press:
Cambridge, Mass
Park, S., H. Piezunka and L. Dahlander (2024), ‘Coevolutionary lock-in in external search’, Academy of
Management Journal, 67(1), 262–288.
Radner, R. (1993), ‘The organization of decentralized information processing’, Econometrica: Journal of the
Econometric Society, 61(5), 1109–1146.
Rogers, E. (2003), Diffusion of Innovations, 5th, Simon and Schuster: New York
Sah, R. K. and J. E. Stiglitz (1985), ‘Human fallibility and economic organization’, The American Economic
Review, 75 (Papers & Proceedings): pp. 292–297.
Sah, R. K. and J. E. Stiglitz (1986), ‘The architecture of economic systems: hierarchies and polyarchies’, American
Economic Review, 76, 716–727.
Sah, R. K. and J. E. Stiglitz (1991), ‘The quality of managers in centralized versus decentralized organizations’,
Quarterly Journal of Economics, 106(1), 289–295.
Schilke, O. (2014), ‘On the contingent value of dynamic capabilities for competitive advantage: the nonlinear
moderating effect of environmental dynamism’, Strategic Management Journal, 35(2), 179–203.
Siggelkow, N. (2007), ‘Persuasion with case studies’, Academy of Management Journal, 50(1), 20–24.
Soh, P. H. and E. B. Roberts (2005), ‘Technology alliances and networks: an external link to research capability’,
IEEE Transactions on Engineering Management, 52(4), 419–428.
Suarez, F. F. and J. M. Utterback (1995), ‘Dominant designs and the survival of rms’, Strategic Management
Journal, 16(6), 415–430.
Tucci, C. L., H. Chesbrough, F. Piller and J. West (2016), ‘When do rms undertake open, collaborative activities?
Introduction to the special section on open innovation and open business models’, Industrial and Corporate
Change, 25(2), 283–288.
Tushman, M. L. (1978), ‘Technical communication in R & D laboratories: the impact of project work
characteristics’, Academy of Management Journal, 21(4), 624–645.
Tushman, M. L. and R. Katz (1980), ‘External communication and project performance: an investigation into
the role of gatekeepers’, Management Science, 26(11), 1071–1085.
Van den Bosch, F. A., H. W. Volberda and M. De Boer (1999), ‘Coevolution of rm absorptive capacity and
knowledge environment: organizational forms and combinative capabilities’, Organization Science, 10(5),
551–568.
Van Zandt, T. (1998), ‘Organizations that process information with an endogenous number of agents’, in M.
Majumdar (ed.), Organizations with Incomplete Information, Cambridge University Press: Cambridge
Volberda, H. W., N. J. Foss and M. A. Lyles (2010), ‘Perspective—Absorbing the concept of absorptive capacity:
how to realize its potential in the organization eld’, Organization Science, 21(4), 931–951.
Winter, S. G. (1987), ‘Knowledge and competence as strategic assets’, in D. J. Teece (ed.), The Competitive
Challenge, Ballinger Press: San Francisco
Yin, R. K. (2009), Case Study Research: Design and Methods, 5th, Sage Publications: London
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025
Internal organization and absorptive capacity in open innovation contexts 23
Zahra, S. A. and G. George (2002), ‘Absorptive capacity: A review, reconceptualization, and extension’, Academy
of Management Review, 27(2), 185–203.
Zobel, A. K., B. Balsmeier and H. Chesbrough (2016), ‘Does patenting help or hinder open innovation? Evidence
from new entrants in the solar industry’, Industrial and Corporate Change, 25(2), 307–331.
Industrial and Corporate Change, 2025, 00, 1–23, DOI: https://doi.org/10.1093/icc/dtaf018, Original Article
© The Author(s) 2025. Published by Oxford University Press in association with the Industrial and Corporate Change
Association.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://
creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any
medium, provided the original work is properly cited.
Downloaded from https://academic.oup.com/icc/advance-article/doi/10.1093/icc/dtaf018/8131777 by guest on 17 May 2025