Innovating in Digital Platforms: An Integrative Approach
Fábio Neves da Rocha and Neil Pollock
University of Edinburgh, Business School, 29 Buccleuch Place, Edinburgh, U.K.
Keywords: Digital Platform, Digital Innovation, Generativity, Generification, Enterprise System, ERP.
Abstract: Increasing competitive pressures are leading companies to innovate through digital platforms. The dominant
theme within extant research on innovation in these platforms conceptualises two different processes:
Generativity (Tilson et al., 2010; Yoo et al., 2012) and generification (Hanseth and Bygstad, 2015; Pollock
et al., 2007). Each of the conceptualisations gives extensive accounts separately, but they have questionable
ability to provide a full understanding of innovation in digital platforms when there is a plural occurrence of
these processes (Sørensen and Williams, 2002). Drawing on an analysis of rich archival data complemented
by interviews reporting five-year relationship between a platform owner and its customer, we revisited
underlying assumptions of its processes. We argue that generativity and generification are related to each
other in a constant flux in which one fuels the other. In this relation, control has new roles other than as key
factor for innovation productivity (cf. Eaton et al., 2015; Yoo et al., 2012), and it is subordinated to the
purpose of innovation. As a consequence, innovation purpose seems to constrain the ‘control vs autonomy’
paradox (Lyytinen et al., 2017).
1 INTRODUCTION
Scholars in Information Systems (IS) argue that
competition today is increasingly promoting
innovation associated with a platform (de Reuver,
Sørensen, and Basole, 2018). Large-scale
‘consumer’ platforms of leading digital companies
“such as Google, Facebook, and Apple” (Yoo 2013)
seem to aim at attracting the largest crowd of
consumers possible by becoming the ‘one-stop shop’
for any possible computing need. The type of
innovation that surrounds these platforms has been
called ‘generativity’ (Tilson et al., 2010; Yoo et al.,
2012). Generativity presents not just innovative and
often disruptive outcomes, but also struggles to
generate them. These conflicts conform a set of
intertwined paradoxes intrinsic to digital platforms
(Lyytinen et al., 2017), which generative
performance depends on the delicate balancing and
reconciliation of these tensions.
In ‘corporate’ digital platforms such as
Enterprise Systems (ES), there is another type of
innovation process, known as ‘generification’
(Hanseth and Bygstad, 2015; Pollock et al., 2007),
which makes digital platforms able to work across
many different contexts, accessing new markets and
new customers.
While sympathetic to these two conceptual
frameworks, we are dissatisfied with their research
approaches: Generativity and generification were
studied as if they occur in isolation. Innovation —
its processes and outcomes — are shaped by the
interrelationships of an array of social and technical
factors that are all configured together (Sørensen
and Williams, 2002). Therefore, as long as there are
evidences that generativity is also found in corporate
digital platforms (e.g. Törmer, 2018) along with
generification, isolated frameworks are doomed to
capture only partial accounts of innovation. We thus
want to know, how do generativity and
generification relate to each other? How are they
performed and what does motivate such
performances?
In this paper, we address these questions through
a case study of a five-year relationship between two
organisations (a platform owner and its customer)
around co-developments in a corporate digital
platform — Enterprise Resource Planning (ERP), a
type of ES. By exploiting rich archival data
complemented by interviews, we could look across
the processes of generification and generativity,
identifying how and why those two firms innovate.
The evidence presented in this article suggests that
there are new roles for ‘control’ other than as key
Neves da Rocha, F. and Pollock, N.
Innovating in Digital Platforms: An Integrative Approach.
DOI: 10.5220/0007711105050515
In Proceedings of the 21st International Conference on Enterprise Information Systems (ICEIS 2019), pages 505-515
ISBN: 978-989-758-372-8
Copyright
c
2019 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
505
factor for innovation productivity (cf. Eaton et al.,
2015; Yoo et al., 2012), in which control is
subordinated to the purpose of innovation. As a
consequence, innovation purpose seems to keep the
‘control vs autonomy’ paradox (Lyytinen et al.,
2017) latent (Smith and Lewis, 2011), expanding our
knowledge on how these paradoxes (not) manifest
themselves. Through linking all these elements, we
derive a model of the circular dynamics of
innovation in large-scale corporate digital platforms
in which generativity and generification work in
tandem and fuel each other.
2 PLATFORM INNOVATION
Organizations are immersed in a world in which its
very fabric is increasingly composed of digital
platforms (Parker et al., 2016; Tiwana, 2014).
Virtually all organizations have some degree of
digital technologies taking care of their operations
(Orlikowski and Scott, 2008), and many have digital
platforms as part of the core of their products and
services (de Reuver et al., 2018). These platforms
are said to produce innovation “that were not
originally imagined by themselves” (Yoo 2013,
p.230).
Digital platform is a concept in development. It
can be seen as a sociotechnical association involving
technical components (e.g. software and hardware)
and related organisational processes and standards
(Tilson et al., 2010) that enable value-creating
interactions between external producers and
consumers (Constantinides et al., 2018). We unpack
innovation processes — generativity and
generification — found in the IS literature of digital
platforms as follow.
2.1 Generativity
The notion of generativity was originally coined by
Zittrain as “a system’s capacity to produce
unanticipated change through unfiltered
contributions from broad and varied audiences”
(Zittrain 2008, p.70). The key examples often
discussed in the literature are of how Google,
Facebook, and Apple deliberately created digital
platforms that were generative, acting as a
‘disrupting force’ (Remneland-Wikhamn et al.,
2011).
Although the original idea suggests rather
chaotic developments, recent accounts of the term
propose that generativity can have a ‘balance of
controls’ (Tilson et al., 2010), or a ‘curation’ (Eaton
et al., 2015; Parker et al., 2016) of its innovation. By
curating the supply of complements, a platform
owner at the same time keeps centralised control
over apps acceptance and facilitates (reasonable)
developer autonomy to innovate (Lyytinen et al.,
2017).
Despite platform owner’s agency, curation
effectiveness is determined by the action and
reaction of other (multiple) actors as well, i.e. the
understanding of the dynamics can only be possible
if all actors are observed simultaneously (Eaton et
al., 2015). Bygstad (2017) argued that generative
technology per se has only ‘latent’ generative
capacity, which is unleashed when collective actions
of agents on their interpretation and use of
technology eventually produce innovation.
Bygstad (2017) also discussed generativity
concept under ‘heavyweight IT’ knowledge regime,
the one that we relate to ‘corporate’ platforms since
it is “enabled by systematic specification and proven
digital technology” and focused “on requirements,
reliability and security” (pp.181-182). Yet these
examples have induced different kinds of insights,
e.g. generativity is very different when the co-
developer is a customer. Through licensing that
grants a right to use (and modify) the platform
(Machal-Fulks and Barnett, 2012), software vendors
make available ‘copies’ of their platforms to
customers, which then become in charge of the
platform management. At firm level of analysis,
generativity seems to be an autonomous, ‘private’
process in which customer acts similarly as software
vendors in selecting and maintaining all add-ons.
Generative components are in general co-developed
in commissioned projects with software vendors
and/or partners (IT consulting firms).
What the literature points to is how generativity
within the organisational context seems to be
substantially different from its initial application in
the Google, Facebook and Apple cases. We
foreground some of these differences through
describing a concept that in many respects sits at the
other end of the spectrum to generativity – that of
‘generification’.
2.2 Generification
The notion of ‘generification’ was a
conceptualisation that sought to show how the same
software platform could be extended and applied for
use by the broadest set of users (Hanseth and
Bygstad, 2015; Pollock et al., 2007). Whilst software
technologies are often born for single organisations
(e.g. Ellingsen and Monteiro, 2012), they can be
ICEIS 2019 - 21st International Conference on Enterprise Information Systems
506
suitable for more than one customer or sector. This
scalability compels software vendors to gather
requirements for the broadest set of possible needs,
e.g. firms in the same industry, and the
interconnection among them (Bygstad, 2017). It
represents an increasing challenge to continuously
evolve specialised industry skills to keep servicing
many markets (Evans et al., 2006) while promoting
an integration of this varied expertise (Tiwana,
2014). To approach this challenge, vendors usually
involve customers and partners early in the design
phase. However, rather than incorporate unfiltered
contributions (cf. Zittrain 2008) they would carefully
select and control which customers (Johnson et al.,
2013), partners (Ceccagnoli et al. 2012; Sarker et al.
2012), and their respective innovations to include in
the actual platform. They chose not only those bits
of functionality that were missing from the platform
and necessary for the new context but those that
would make the system more ‘global’ (e.g. attractive
to the largest number of existing and potential users
in any geography).
Likewise, in contrast to the idea that products
and services would move in direction not originally
imagined by themselves (cf. Yoo, 2013, p.230), the
vendor would make strategic choices about which
sectors and industries they wished to enter. In
travelling to a new sector, the platform would be
incomplete by design (Scott and Kaindl, 2000).
Being global means capturing the most common
functionalities for the targeted market, leaving
idiosyncratic demands behind. Some abandoned
functionalities may be considered compulsory to
some customers, opening room for procurement of
alternative, off-the-shelf solutions (Light et al.,
2001), or for ‘home-made’ developments. One way
or the other, generification process would begin
again, selecting further user innovations that would
help it to move to the next user, sector, industry, etc.
(Pollock and Williams, 2009).
Generification is not unproblematic (cf.
Ellingsen and Monteiro, 2012; Pollock et al., 2007),
nor is generativity (cf. Eaton et al., 2015;
Remneland-Wikhamn et al., 2011). The performance
of both processes is surrounded by struggles and
conflicts, which can be viewed as ‘paradoxes’.
2.3 Paradoxes
Over the last decade IS scholars have so far revealed
a great concern on issues related to the expansion
(heavily based on innovation) and control of digital
platforms (e.g. Boudreau 2012; Hanseth and
Bygstad 2015), identifying some of them as
paradoxes (e.g. Tilson et al. 2010; Wareham et al.,
2014; Zittrain, 2008). Lyytinen et al. (2017)
acknowledged the importance of the issues related to
the expansion and control of digital platforms. The
authors argued these issues are part of a more
complex interaction of four different paradoxes
entwined and mutually constitutive of the dynamics
of digital platforms: (1) fixity vs variety, (2) stability
vs change, (3) local vs global, and (4) control vs
autonomy. The last one seems to be especially
relevant to our research here.
Control vs autonomy paradox presents
conflicting forces opposing centralised and
distributed control (or individual autonomy).
Empirical studies on this matter (e.g. Eaton et al.,
2015; Lyytinen et al., 2017; Tilson et al., 2010) had
consumer digital platforms as their settings, and
affirmed that control is a key factor for generativity
performance. Authors argued that, on one hand,
many innovative apps may not survive platform
owner’s curation, and the more it gets restrictive, the
more it discourage further developments (Eaton et
al. 2015; Yoo et al. 2012). On the other hand, less
restrictive curation may allow low-quality or even
malicious apps to enter (Eaton et al. 2011), leading
to negative customer experience and potentially
harming platform’s reputation and economic
sustainability (Boudreau 2012).
It is not clear to what extent this paradox can be
seen in corporate digital platforms. We will examine
these characteristics further with empirical material.
3 RESEARCH CONTEXT AND
METHODS
The following is a single instrumental case (Stake,
1995) in which an ERP is crafted by a global
software vendor and a large British university. The
case was chosen because it involves fundamental
elements of the studied phenomenon (cf. Yin 1994):
both generification and generativity innovation
processes are present. The data are archival,
composed of 2,000 e-mails, contracts, business
presentations, along with information from the
websites of the co-development actors, other ERP
vendors, industry analysts and specialised IT media,
all collected over five years. Eight additional
interviews with key actors (operational and C-
levels), ranging from 29 to 137 minutes with an
average length of 68 minutes, were recorded,
transcribed and added to the body of data.
Innovating in Digital Platforms: An Integrative Approach
507
The data, then, was inductively analysed
following the assumptions of grounded theorising
(Eisenhardt, 1989; Glaser and Strauss, 1967) and
continually compared with insights from the
literature. In vivo phrases and terms mentioned by
informants produced first-order codes (Van Maanen,
1979) with the use of NVivo 12, a qualitative data
analysis software. The goal was to understand what
actors did in their interactions for technology co-
development and how they justified them.
The second round of coding was a result of the
comparison of in vivo codes with both themselves,
to underline emerging concepts and their
interrelationships. At this point, we also derive the
emergent categories by comparing our data with
existing theoretical frameworks. This favoured a
consolidation of some in vivo categories into a set of
second-order notions. This included identifying
various ways whereby co-development activities are
performed and the motivation underlying such
performances.
As we recognised links and interrelations
between second-order categories, we could
consolidate these into larger groupings, which
conforms our insights at a more abstract level (Ryan
and Bernard, 2003). This more logically ordered set
of categories thus obtained focused on how
organisations manage the innovation in corporate-
type platform. This finally led us to identifying as
overarching themes the controlling mechanisms and
the purposes of the innovation. The final coding
structure for each innovation process is brought by
Table 1 (the controlling mechanisms) and Table 2
(the purposes).
Table 1: Controlling Mechanisms - What Actors Do.
First Order
Second Order
Overarching Theme
GENERATIVITY
“design, specification, templating, documentation”, “workshops, notes,
sign-off, specifications”, “an idea of the scale of the work”, “Blueprint”,
“scope for the project”
definition of
innovation scope
controlling mechanisms
of innovation processes
in corporate digital
platforms
“The HE market is particularly difficult for a generic software house to
crack”, “most high-profile of the suppliers around”, “market leader”,
“address our business problem”
identification and
allocation of suitable
resources
“symbiotic partnership”, “cannot accept the revised wording”, “do not
see the same scope for the project”, “Blueprint is now 4 weeks later than
originally planned”, “The financial projections for the project have gone
horrendously awry”, “polarised state”, “transfer our knowledge,
documentation and other resources”, “fighting to keep the costs down”,
“cutting out functionality to meet a specific price”,
“developed/implemented”
resource
orchestration and
conflict management
“not only in the UK, but also in Europe and probably worldwide”, “a lot
of firsts”, “70% overall discount”, “reduce the cost of the over-run on the
project”, “Lower total cost of ownership”, “really attractive product”, “
revenue stream”, “change their business”, “making a value out of an IT
implementation”, “something which creates a real win-win”, “superb
solution to a vitally important issue which faces all Universities”
identification and
exploitation of
benefits emerged
from co-development
GENERIFICATION
“methodology in … software development”, “you do a plan for that
product”, “scenario descriptions”, “use cases”, “The specification
describes how the software should work”, “Depending on the
requirement”, “customer requirement”, “verify what you know about the
process”, “to put something in the standard product, it must be so much
more general, must be customise-able, there must be some configuration
options in it”, “community meetings”
definition of
innovation scope
“certain sense of cooperations”, “a bit visionary”, “leader or early
adopter”, “experts on the subject”, “somebody who is kind of in the
mainstream”
identification and
allocation of suitable
resources
“sense of cooperation”, “partnership”, “overlaps with your interest”,
“committed”, “too costly”, “not manageable”, “software costs”,
“implementation costs”, “development partners”, “need our support”,
“idiosyncrasies could be accommodated”, “being able to influence”
resource
orchestration and
conflict management
“business case”, “strong reference”, “intended to be a global product”,
“entering the HE market globally”, “win new customers”, “they want to
sell it more”, “make money on it”, “there’s a need for a product”, “value
of the development”, “as flexible as possible”, “functionality can be
configured in more than one way”
identification and
exploitation of
benefits emerged
from co-development
ICEIS 2019 - 21st International Conference on Enterprise Information Systems
508
Table 2: Innovation Purpose - Why Actors Do What They Do.
First Order
Overarching Theme
GENERATIVITY
“commercialisation to mutual benefit”, “toss in the MBA item”, “potential
‘win bonuses’ or ‘payments’”, “becoming a ‘Centre of Excellence’”,
“shared financial rewards”, “we receive £1,000 per visit for this”
purposes for co-
innovating in corporate
digital platforms
“role as a ‘partner’ in the ‘development’ of leading edge . . . applications”,
“if we block their development in the HE sector them we are doing a dis-
service to the community...”, “Best Public Sector Project”, “improve our
position as a reference site”
“address the most critical Business issue within the University”. “maximise
the benefits”, “The purpose: to drive growth and enhance its competitive
edge”, “very best back-office systems (to support students, researchers and
administrators) in the world”, “the VALUE which the . . . solution is
bringing to our recruitment Income stream”, “This has a real value to a
seller”
GENERIFICATION
“able to influence campus management”, “policy meetings and practition
meetings”, “every major UK University is very aware of what we are
doing!”, “we continue to enhance our reputation as one of the United
Kingdom's leading educational institutions”, “the first of a very small
number of global Universities to become ‘development partners’”
“it brings in revenue”, “leverage our investment . . . to support our strategy
for profitable growth”, “it needed to replace its Student Record system”,
“The new integrated solution covers nearly all student administration and
accounting processes”, “replaces a number of disparate in-house and niche
products, and their associated interfaces”
4 FINDINGS
Universities are no different from other large
organisations. They have complex operations and
the adoption of a corporate digital platform – an
ERP – was the chosen alternative for many of them
to confront this complexity. It was the case of a large
British university, UniBrit, which selected their ERP
from GlobalSw, a global leader in corporate
platforms. Until then, GlobalSw’s market leadership
was reached in sectors other than Higher Education,
and so the vendor had signalled intent to invest in
developing its software to meet specific needs of this
new market.
Short after, during the implementation of
UniBrit’s ERP, the university was invited by
GlobalSw to take part in the development of the
Higher Education System (HES) as a ‘developer
partner’ in a small group composed of analysts and
programmers from GlobalSw and few other large,
famous universities from diverse countries. UniBrit
engaged quickly and was particularly active in the
development, contributing materially and critically.
HES was meant to be the complimentary system
specific for campus management that would be
seamlessly integrated into the ERP platform.
Therefore, it comprised not only developing new
functionalities but also tailoring some of existing
modules in a way to meet new requirements and
provide necessary integration.
4.1 Recipe for Generification:
Recycling, Creating, Failing,
Trying Again
A new development does not necessarily starts from
scratch. GlobalSw has grown its software portfolio
overtime largely enough to employ its components
as recyclable building blocks of new products. A
GlobalSw’s developer explained the relevance of
recycling in new system developments:
… we have our subsidiary’s sales organisations
and we have industry solution management, which
[manages] 27 industry solutions, one is Higher
Education and Research. … [W]e have customers
requiring software here, and industry solution
management tries to fit the different products
together or to say, here we have a blank space, we
should develop a product … And they do the gap
analysis and [confirm the necessity of development],
or sometimes it’s only a reconfiguration required.
So, industry solution management is really linking
the two together [, recycling and new development].
(interview, GlobalSw’s Developer)
GlobalSw’s ‘gap analysis’ pointed out several
modules (originally developed for large corporations
of other industries) that could be recycled for the
project. However, they were rejected by the
developer partners. The Real Estate Management
module, for instance, was scrutinized by UniBrit. In
an internal memo, it was considered so “far
Innovating in Digital Platforms: An Integrative Approach
509
removed” from the university’s needs the suggestion
was to “start again from scratch than try to adjust
the existing module.” Most of the work was, then, in
creating the new pieces specific to Higher
Education. Despite the small number of participants
in the development group, the effort was by no
means straightforward. For instance, any of the
developer partners could individually suggest a
change in the system (e.g. inclusion of a
functionality), which would circulate back and forth
in a closed network of participants under the
coordination of GlobalSw, who wanted to make
sense of each of the suggestions to find those that
are as generic as possible. The following example is
about the business process known as ‘progression’,
which was intended to measure students’
performance over time. The vendor tried to
consolidate different processes into a single, global
business process with variants that could keep major
national idiosyncrasies. However, it did not end up
as expected, according to a GlobalSw’s developer:
… we realised the US model of progression and
the UK-based model of progression was very
different. … And that took a bit of time for us to
realise that we shouldn’t build one process with two
variants but really we should build two different
processes. … Now we have what we call a
programme type progression, which is used by US
universities, and we have a programme progression,
which is stage based, which is mainly for UK
universities… (interview, GlobalSw’s Developer)
The software vendor had to give up from having
one single progression process. Instead, GlobalSw
allowed the coexistence of different processes
separately.
4.2 When Generification Fails…
Generativity!
In the next two years, the HES system was fully
developed, tested at UniBrit’s premises, and then
became operational in the university. A formal
contract related to the HES system, called licensing
agreement, was signed. It stated general grounds of
the relationship between GlobalSw and UniBrit
about uses, copyright, and liabilities. Specifically in
terms of co-development, the contract allowed
UniBrit to create and/or modify the systems that
compose the platform. For doing so, UniBrit needed
a special code key from GlobalSw, with which the
vendor identifies the author of any update in the
platform.
The contract determined the intellectual property
rights (IPR) of the creations and changes made by
UniBrit shall belong to the university but subject to
the following restrictions:
[UniBrit] shall only be entitled to assign the
Intellectual Property Rights in such modifications to
third parties after first obtaining the written consent
of [GlobalSw] (such consent not to be unreasonable
withheld).
[GlobalSw] may at any time demand the
assignment to it of all Intellectual Property Rights in
such modifications (or any of them) in return for the
payment by [GlobalSw] to [UniBrit] of suitable
compensation in which case [GlobalSw] will grant
to [UniBrit] an irrevocable royalty-free license-back
to use the Intellectual Property Rights in such
modifications. (licensing agreement clauses)
UniBrit used HES extensively during the first
year of operation. The university had an ambitious
growth plan for its post-graduation courses with
international students. However, UniBrit failed to
achieve the projected volume, causing a huge
financial impact. The failure was rooted in poor
hiring processes (still fully manual). UniBrit, then,
started procuring systems that could add those
functionalities to HES. Two firms came to the short
list: an American software vendor and GlobalSw.
The American vendor’s proposition was technically
superior, but the balance eventually fell to GlobalSw
due to the strength of its influence:
There was considerable pressure from within the
University to adopt a third party product, [American
vendor’s product], but I managed to persuade my
colleagues on our Executive Board to stick with
[GlobalSw] on the understanding that you were fully
committed to the project. (e-mail from UniBrit’s
Deputy Vice Chancellor to GlobalSw’s Business
Development Director)
Different from traditional licensing when the
software is already developed and ready for use,
custom development project is a service paid by
customer as project advances. Therefore, the project
was grounded in a different type of contract, which
one of the clauses originally stated that:
All intellectual property rights arising from the
provision of the Services shall automatically vest in
[GlobalSw] or its licensors and [UniBrit]
undertakes to execute such documentation as may be
necessary to perfect the title of [GlobalSw] (or its
licensers) to such rights. (services contract clause)
The signed version, however, did not have such
clause. It was suppressed by UniBrit during the
contract negotiation and GlobalSw did not notice
that. By abolishing it, the university hoped they
could own the IPR stemmed from the development,
ICEIS 2019 - 21st International Conference on Enterprise Information Systems
510
as stated in the previous (and still valid) licensing
agreement contract.
Because “[GlobalSw] had misunderstood the
size of the gap between what [UniBrit]’s business
requirements stated and what their standard
software modules could deliver” (UniBrit’s CIO
internal e-mail), the costs of the project grew
dramatically. GlobalSw charged more because it had
to employ additional resources to develop several
functionalities that were considered for granted
(recyclable) at the outset. These resources came
from a commissioned third party: SwServices, a
consulting firm specialised in Human Resources that
was GlobalSw’s partner.
4.3 Custom Development Opens Room
for Generification
UniBrit was very concerned about the participation
of a third party in the development because of a
potential leak of IPR, since SwServices was not
legally entitled to that IPR. The university’s concern
was based on the market practise (cf. Wareham et
al., 2014). Consulting firms like SwServices usually
lead custom developments by themselves, which
resulting software eventually becomes a product that
is sold to other customers, just like GlobalSw does
with their own products. Some are even marketed
and sold by software vendors themselves, as an
executive of GlobalSw explained:
… consulting companies … they make … their
own IP, their own development an important part of
their business. So saying that instead of developing
one by one to each customer, why not to add pieces
of software to the standard [GlobalSw’s] software?
… [T]hey can transform it also in a product to sell
to the market. Of course to sell it … in conjunction,
adding it as a component, a kind of add-on to the …
technology vendor’s software. (interview,
GlobalSw’s VP of Customer Experience)
There was no evidence that SwServices had
developed similar software based on non-authorised
information from UniBrit’s project. But it was clear
that GlobalSw generified co-created complements.
The following e-mail was sent from GlobalSw
asking UniBrit permission for sharing some
developments of the project with an American
university that was also a customer of the software
vendor (in another development project). The
software vendor justified its claim by arguing that
UniBrit’s development was already benefited by
some information from that American university:
[The American university] … has been very
helpful during the blueprint phase providing me
information about their admission process workflow,
which I have also shared with [UniBrit] project
team and taken as input for the blueprinting and
further process. In return I would like to send [them]
some of the functional and technical specifications
as they are thinking about enhancements to their
process. … Would be o.k. from your side? (e-mail
from a GlobalSw’s Developer to UniBrit’s Project
Manager)
A GlobalSw developer briefly explained the
future generification of scanning and management
functionality:
… If I notice, okay, this requirement hasn’t been
considered in our standard product … I kind of feed
it back for the future. … [UniBrit] now attaches
documents to applications, like this document
scanning piece, this is something we look at in our
next release. (interview, GlobalSw’s Developer)
Meanwhile, UniBrit’s CIO was trying to seek
alternatives to finance the continuity of the project,
since the costs were going to exceed the budget.
Because of the software’s commercial potential, the
executive thought it would be a good idea to sell
university’s IPR to GlobalSw as a deduction of
project costs. The negotiation about cost reduction
took several months and eventually led the vendor to
lower them a bit. During the negotiation, UniBrit
was forced to sign an additional contract, which
replaced the first one and had back the (missing)
clause securing all IPR to GlobalSw, effective from
the signature date onwards. On the other hand,
GlobalSw agreed to pay UniBrit per its services as
‘reference actor’ (Pollock and Hyysalo, 2014) each
time the university hosted visitors interested in the
system. In the short term, these payments
represented only a small relief on the project’s costs.
But they were approached in a different way by both
firms, as a long-term ‘partnership’:
As we discussed, let's plan on you and [UniBrit’s
Dean] meeting with us (myself, [VP of
Development], and [Development Director]) in
Paris around the Advisory Council meeting,
depending on travel schedules. This will give us a
chance to discuss in detail how we can work
together in the coming months and years to achieve
our common goals. … I Will go ahead and give the
“green light” to the visit from the Norwegian
customer … based on our standard reference
bonus… (e-mail from GlobalSw’s Business
Development Director to UniBrit’s CIO)
I have maintained throughout this project that
this must be viewed as a “Partnership” between
[GlobalSw] and [UniBrit]. [UniBrit] positively
wants [GlobalSw] to become increasingly successful
Innovating in Digital Platforms: An Integrative Approach
511
across the Higher Education sector both in the UK
and elsewhere. (e-mail from UniBrit’s CIO to
GlobalSw’s Head of Consulting)
For GlobalSw the partnership meant business
development led by an expert, UniBrit. For the
university, the partnership meant at the same time a
new source of revenue and, most importantly, an
enhancement of university’s reputation in its market:
We will leverage our investment in [GlobalSw]
software to support our strategy for profitable
growth as we continue to enhance our reputation as
one of the United Kingdom's leading educational
institutions. (UniBrit’s Assistant Director or
Information Systems and Services in a GlobalSw’s
advertisement material)
You may be aware that we expect to have a very
large number of [visitors] in October (ie every
major UK University is very aware of what we are
doing!...) (e-mail from UniBrit’s CIO to GlobalSw’s
Services Director)
Despite the relative success in exploring new
opportunities that emerged from co-development,
the negotiation about the value of the IPR entitled to
both firms (supported by the first co-development
contract) and how to realise it did not settle when the
data collection for the case was finished.
5 DISCUSSION AND
CONCLUSIONS
In order to stay competitive, organisations today are
increasingly promoting innovation associated with a
platform (de Reuver et al., 2018) that eventually
produce unplanned products and services (Yoo
2013, p.230) through an innovation process known
as generativity (Tilson et al., 2010; Yoo et al., 2012).
Studies of leading consumer platform have
sharpened our understanding of not just the
innovative and often disruptive outcomes, but also
paradoxes faced whilst generating them (Lyytinen et
al., 2017). On corporate platforms, the studies have
shown how platform leaders travel their platforms to
new contexts by performing a different innovation
process, called generification (Hanseth and Bygstad,
2015; Pollock and Williams, 2008; Pollock et al.,
2007). Generification is also said to face struggles
during its course.
These conceptualisations worked well enough
when only one innovation process is performed in a
given digital platform, but fall short when there are
multiple innovation processes occurring in that
setting (Sørensen and Williams, 2002). We argue
that more recent evidence is beginning to question
previous knowledge on digital platform innovation,
inviting us to revisit underlying assumptions of its
processes.
5.1 Innovation in the Digitised World
Revisited
Evidences from our empirical case showed
generativity and generification are not as
straightforward as one might imagine. That is,
generativity goes hand in hand with generification.
Let us first summarise the two processes as distinct
and then reflect how they work in tandem.
5.1.1 Generification in Corporate Platforms
Digital platforms, we argue, are purposeful artefacts
(Rosenblueth et al. 1943), having goals to be
attained attributed by platform owners and users.
Large-scale corporate platforms are strictly focused
on business management. Generification brings
together actors that have clear orientation towards
the achievement of organisational goals and increase
of influence and reputation. In their ambition to
extend platform’s reach to new markets (Pollock et
al., 2007), platform owners attract influent
customers for co-development, hoping to add both
consistent specialised knowledge and reputation
(Ravasi et al., 2018), i.e. signalling their new
strengths to the market. Customers, in turn, look for
return on investment, i.e. cope with the business
benefits the platform promised to deliver, and also
try to leverage their reputation (ibid) over platform
owners and other players in their market.
Platform owners purposefully control
generification by selecting who will take part in the
co-development (Ceccagnoli et al., 2012; Johnson et
al., 2013; Pollock et al., 2007; Sarker et al., 2012)
and which innovations will be part of the platform.
Boundary resources (Ghazawneh and Henfridsson,
2013) like contracts and code keys are used to
control how both generification and (potential)
generativity unfold. However, this closer control
does not make the development unambiguous. For
instance, recycling failed to accommodate the
(specialised) requirements of the new industry, and
co-created piece had to be ‘de-generified,’ allowing
coexistence of multiple particular templates (Pollock
et al., 2007).
ICEIS 2019 - 21st International Conference on Enterprise Information Systems
512
5.1.2 Generativity in Corporate Platforms
Generativity in large-scale corporate platforms is
closer to earlier software traditions of bespoke
development. Customers’ objective is to fill critical
gaps of products and services caused by platform
incompleteness (Scott and Kaindl, 2000), i.e.
generification, with personalised pieces that address
their idiosyncratic needs, making the platform more
specialised. In this way, they lead generative co-
developments that encompass the same core
motivation found in the generification process
(achievement of business goals and reputation
increase). The level of control is similar to
generification as well (selection of partner,
innovation definition, contracts and development
keys). Interestingly, tight control, boundary
resources, and asymmetric relationship did not
discourage customer to lead a generative venture.
Generativity, in our case, was fuelled by customer
necessity of complementing the platform with key
functionalities for its operations, suggesting that
‘control’ may have new roles.
5.1.3 The New Roles of Control
IS literature (e.g. Eaton et al., 2015; Tilson et al.,
2010; Wareham et al., 2014; Yoo et al., 2012)
presented generativity and control as entities that
seat at different ends of a see-saw: higher control
leads to lower generativity and lower control allows
higher generativity. Lyytinen et al. (2017)
proclaimed the paradox of centralised and
distributed control (or individual autonomy) is one
of the four paradoxes that compose the emergent
property of generativity.
We see it differently. In corporate platforms
generativity and control are not inversely
proportional. Control can be found in a more fine-
grained characterisation: inside the innovation
process. There, control is part of the very fabric of
generativity. It starts by defining the scope of what,
how and when is to be jointly developed along with
who becomes entitled for which value created. The
innovation definition is usually inscribed in the
format of contracts and blueprints. It is then
followed by resources allocation, when necessary
components for the co-innovation venture are
identified and mobilised. After allocation, resources
are managed to produce innovation. The co-
development supervision involves the build-up of the
planned generative component with planning
reviews and conflict management along the way. At
the end participants assess benefits emerged from
the co-development venture, which includes but is
not restricted to IPR, and (try to) appropriate them.
Innovation definition, resources allocation, co-
development and value appropriation are controlling
mechanisms found in both innovation processes
(generativity and generification), which we define as
the micro-foundations of large-scale corporate
platform innovation.
Additionally, we saw how influence played a
decisive role in the whole generative process. There
was a notable influence asymmetry between the
platform owner and the customer (cf. Eaton et al.,
2015). When platform owners become one of the
few large and resourceful vendors, customers may
find more attractive to collaborate with them other
than competitive alternatives in the ecosystem.
However, generative ventures with influential
vendors may lead to one-sided abuse (Eaton et al.,
2015). Since platform owners’ priorities “are to
protect their own interests and secure their
competitive positions” (Constantinides et al., 2018,
p.4), boundary resources (Ghazawneh and
Henfridsson, 2013) such as contracts may be
designed and enforced in a way that vendor’s
interests are primarily (or solely) contemplated.
Even when co-authors retain rights of authorship,
“[t]here is strong precedent for platform [owners] to
appropriate developer innovations” and “make these
features available to the entire market” (Parker and
Van Alstyne, 2018, p.3016), i.e. generifying the
generative complement.
Incompleteness of generified platform leads to
generativity, which in turn can be generified. We
argue that corporate platform innovation follows a
cycle, a moto continuum in which generativity and
generification are entwined and successively fuel
each other. Figure 1 shows the dynamics of large-
scale corporate platform innovation.
We can now draw a picture for scholars
interested in digital platform innovation by
reflecting on the implications of the different kinds
of innovation that occur in digital platforms.
5.2 Conclusions and Future Directions
Yoo et al. (2012), Eaton et al. (2015) and Lyytinen
et al. (2017) asserted that organisations should
structure themselves to oversee the delicate and
paradoxical balance of generativity and control in
the platform. We see some aspects in a different
way.
Innovating in Digital Platforms: An Integrative Approach
513
Figure 1: The Innovation Processes Dynamics in Large-
Scale Corporate Platforms.
We suggest that platform innovation produces
different outcomes depending on the purpose of the
platform rather than control. Purpose is the basis for
(strategic) actions (Mintzberg, 1987), and a strategy
is required to enable the ongoing innovation
management (Dougherty and Dunne, 2011).
Different purposes may lead, we argue, to different
innovation dynamics. Therefore, rather than control,
it is the purpose that orientates strategy that, for its
turn, encourages innovation to walk unimagined
paths, e.g. in iOS and Android platforms, or to
(roughly) follow a road map, e.g. in an ERP
platform, since many gaps are known.
Consequently, we acquire a window onto the
view of new roles of control in digital platforms. We
see control serving as strategy enabler, having
boundary resources (Ghazawneh and Henfridsson,
2013) as instruments for its implementation,
ordering how innovation processes unfold. The four
micro-foundational mechanisms of platform
innovation we have found – innovation definition,
resource allocation, co-development and value
appropriation – can serve as a foreground, we hope,
to help the understanding of innovation governance,
especially because innovation is often problematic,
as we saw in our empirical case and is supported by
the extant of IS literature (e.g. Lyytinen et al., 2017;
Pollock et al., 2007). These certainly do not
constitute the full pack of platform innovation
mechanisms, but they can serve as steppingstone for
a further investigation on, for instance, the practices
(e.g. Vaast and Walsham 2005) around these
processes, how actors perform them, and respective
interaction with platform governance at higher levels
(e.g. Tiwana et al., 2010).
ACKNOWLEDGEMENTS
We would like to thank Carsten Sørensen, Eric
Monteiro, Raluca Bunduchi and Hajar Mozaffar for
their invaluable contributions while redrafting this
paper.
REFERENCES
Boudreau, K. J. (2012). Let a Thousand Flowers Bloom?
An Early Look at Large Numbers of Software App
Developers and Patterns of Innovation. Organization
Science, 23(5), 1409–1427.
Bygstad, B. (2017). Generative innovation: a comparison
of lightweight and heavyweight IT. Journal of
Information Technology, 32(2), 180–193.
Ceccagnoli, M., Forman, C., Huang, P., and Wu, D. J.
(2012). Cocreation of Value in a Platform Ecosystem!
The Case of Enterprise Software. MIS Quarterly,
36(1), 263–290.
Constantinides, P., Henfridsson, O., and Parker, G. G.
(2018). Platforms and Infrastructures in the Digital
Age. Information Systems Research, 29(2), 381–400.
de Reuver, M., Sørensen, C., and Basole, R. C. (2018).
The digital platform: a research agenda. Journal of
Information Technology, 33(2), 124–135.
Dougherty, D., and Dunne, D. D. (2011). Organizing
Ecologies of Complex Innovation. Organization
Science, 22(5), 1214–1223.
Eaton, B., Elaluf-Calderwood, S., Sørensen, C., and Yoo,
Y. (2011). Dynamic structures of control and
generativity in digital ecosystem service innovation:
the cases of the Apple and Google mobile app stores
(No. 183). Information Systems and Innovation Group
(Vol. 44). London.
Eaton, B., Elaluf-Calderwood, S., Sørensen, C., and Yoo,
Y. (2015). Distributed tuning of boundary resources:
the case of Apple’s iOS service system. MIS
Quarterly, 39(1), 217–244.
Eisenhardt, K. M. (1989). Building Theories from Case
Study Research. Academy of Management, 14(4),
532–550.
Ellingsen, G., and Monteiro, E. (2012). Electronic patient
record development in Norway: The case for an
evolutionary strategy. Health Policy and Technology,
1(1), 16–21.
Evans, D. S., Hagiu, A., and Schmalensee, R. (2006).
Invisible engines: how software platforms drive
innovation and transform industries. The MIT Press.
Cambridge: MIT Press.
Ghazawneh, A., and Henfridsson, O. (2013). Balancing
platform control and external contribution in third-
party development: The boundary resources model.
Information Systems Journal, 23(2), 173–192.
Glaser, B. G., and Strauss, A. L. (1967). The Discovery of
Grounded Theory: Strategies for Qualitative
Research. New York: Aldine.
ICEIS 2019 - 21st International Conference on Enterprise Information Systems
514
Hanseth, O., and Bygstad, B. (2015). Flexible
generification: ICT standardization strategies and
service innovation in health care. European Journal of
Information Systems, 24(1), 645–663.
Johnson, M., Mozaffar, H., Campagnolo, G. M., Hyysalo,
S., Pollock, N., and Williams, R. (2013). The managed
prosumer: Evolving Knowledge Strategies in the
Design of Information Infrastructures. Information,
Communication & Society, 17(7), 795–813.
Light, B., Holland, C. P., and Wills, K. (2001). ERP and
best of breed: A comparative analysis. Business
Process Management Journal, 7(3), 216–224.
Lyytinen, K., Sørensen, C., and Tilson, D. (2017).
Generativity in Digital Infrastructures: A Research
Note. In R. D. Galliers and M.-K. Stein (Eds.), The
Routledge Companion to Management Information
Systems (pp. 253–275). Abingdon: Routledge.
Machal-Fulks, J., and Barnett, C. (2012). Enterprise
Software Licensing: New Options - New Obligations.
Texas Wesleyan Law Review, 18(4), 753–765.
Mintzberg, H. (1987). The Strategy Concept I - Five Ps for
Strategy. California Management Review, 30(1), 11–
24.
Orlikowski, W. J., and Scott, S. V. (2008). Challenging
the Separation of Technology, Work and Organization.
The Academy of Management Annals, 2(1), 433–474.
Parker, G. G., and Van Alstyne, M. W. (2018). Innovation,
Openness, and Platform Control. Management
Science, 64(7), 3015–3032.
Parker, G. G., Van Alstyne, M. W., and Choudary, S. P.
(2016). Platform Revolution. How Networked Markets
are Transforming the Economy - and How to Make
Them Work for You. New York: W.W. Norton &
Company.
Pollock, N., and Hyysalo, S. (2014). The Business of
Being a User: The Role of the Reference Actor in
Shaping Packaged Enterprise System Acquisition and
Development. MIS Quarterly, 38(2), 473–496.
Pollock, N., and Williams, R. (2008). Software and
Organisations. Software and organisations: The
biography of the enterprise-wide system or how SAP
conquered the world. London: Routledge.
Pollock, N., and Williams, R. (2009). The sociology of a
market analysis tool: How industry analysts sort
vendors and organize markets. Information and
Organization, 19(2), 129–151.
Pollock, N., Williams, R., and D’Adderio, L. (2007).
Global Software and its Provenance: Generification
Work in the Production of Organizational Software
Packages. Social Studies of Science, 37(2), 254–280.
Ravasi, D., Rindova, V., Etter, M., and Cornelissen, J.
(2018). The formation of organizational reputation.
Academy of Management Annals, 12(2), 574–599.
Remneland-Wikhamn, B., Ljungberg, J., Bergquist, M.,
and Kuschel, J. (2011). Open Innovation, Generativity
and the Supplier As Peer: the Case of Iphone and
Android. International Journal of Innovation
Management, 15(01), 205–230.
Rosenblueth, A., Wiener, N., and Bigelow, J. (1943).
Behavior, Purpose and Teleology. Philosophy of
Science, 10(1), 18–24.
Ryan, G. W., and Bernard, H. R. (2003). Techniques to
Identify Themes. Field Methods, 15(1), 85–109.
Sarker, S., Sarker, S., Sahaym, A., and Bjørn-Andersen,
N. (2012). Exploring Value Cocreation in
Relationships Between an ERP Vendor and its
Partners: A Revelatory Case Study. MIS Quarterly,
36(1), 317–338.
Scott, J. E., and Kaindl, L. (2000). Enhacing functionality
in an enterprise software package. Information &
Management, 37(3), 111–122.
Smith, W. K., and Lewis, M. W. (2011). Toward a theory
of paradox: A dynamic equilibrium model of
organizing. Academy of Management Review, 36(2),
381–403.
Sørensen, K. H., and Williams, R. (2002). Shaping
Technology, Guiding Policy: Concepts, Spaces, and
Tools. Technological Forecasting and Social Change.
Cheltenham: Elgar.
Stake, R. E. (1995). The art of case study research.
Thousand Oaks: SAGE Publications.
Tilson, D., Lyytinen, K., and Sørensen, C. (2010). Digital
infrastructures: The missing IS research agenda.
Information Systems Research, 21(4), 748–759.
Tiwana, A. (2014). Platform ecosystems: aligning
architecture, governance, and strategy. Waltham:
Elsevier Inc.
Tiwana, A., Konsynski, B., and Bush, A. A. (2010).
Platform Evolution: Coevolution of Platform
Architecture, Governance, and Environmental
Dynamics. Information Systems Research, 21(4), 675–
687.
Törmer, R. L. (2018). Internal Digital Platforms and
Generative Mechanisms of Digital Innovation. In
Proceedings of the 39th International Conference on
Information Systems - ICIS 2018 (pp. 1–17).
Vaast, E., and Walsham, G. (2005). Representations and
actions: The transformation of work practices with IT
use. Information and Organization, 15(1), 65–89.
Van Maanen, J. (1979). The Fact of Fiction in
Organizational Ethnography. Administrative Science
Quarterly, 24(4), 539–550.
Wareham, J., Fox, P. B., and Cano Giner, J. L. (2014).
Technology Ecosystem Governance. Organization
Science, 25(4), 1195–1215.
Yin, R. K. (1994). Case Study Research (2nd ed.).
Thousand Oaks: SAGE Publications.
Yoo, Y. (2013). The Tables Have Turned: How Can the
Information Systems Field Contribute to Technology
and Innovation Mangement Research. Journal of the
Assciation for Information Systems, 14(5), 227–236.
Yoo, Y., Boland, R. J., Lyytinen, K., and Majchrzak, A.
(2012). Organizing for Innovation in the Digitized
World. Organization Science, 23(5), 1398–1408.
Zittrain, J. L. (2008). The Future of the Internet and How
To Stop It. New Haven: Yale University Press.
Innovating in Digital Platforms: An Integrative Approach
515