Internet Histories

Digital Technology, Culture and Society

ISSN: 2470-1475 (Print) 2470-1483 (Online) Journal homepage: https://www.tandfonline.com/loi/rint20

Facebook’s evolution: development of a platform-

as-infrastructure

Anne Helmond, David B. Nieborg & Fernando N. van der Vlist

To cite this article: Anne Helmond, David B. Nieborg & Fernando N. van der Vlist (2019)
Facebook’s evolution: development of a platform-as-infrastructure, Internet Histories, 3:2, 123-146,
DOI: 10.1080/24701475.2019.1593667

To link to this article: https://doi.org/10.1080/24701475.2019.1593667

© 2019 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 01 May 2019.

Submit your article to this journal

Article views: 4870

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rint20
INTERNET HISTORIES
2019, VOL. 3, NO. 2, 123–146
https://doi.org/10.1080/24701475.2019.1593667

Facebook’s evolution: development of a platform-as-

infrastructure
Anne Helmonda , David B. Nieborgb and Fernando N. van der Vlistc
a
University of Amsterdam, Amsterdam, the Netherlands; bUniversity of Toronto, Toronto, Canada;
c
University of Siegen, Siegen, Germany

ABSTRACT ARTICLE HISTORY

The purpose of this article is to operationalise an evolutionary per- Received 7 December 2018
spective on the history of social media and to trace Facebook’s evo- Revised 19 February 2019
lution from a social networking site to a “platform-as-infrastructure”. Accepted 8 March 2019
Social media platforms such as Facebook change constantly on the
KEYWORDS
level of their platform architectures, interfaces, governance frame-
Facebook; platform
works, and control mechanisms, all while responding to their larger evolution; platform history;
environments. By examining the evolution of Facebook’s program- programmability; infrastruc-
mability and corporate partnerships, we develop an empirical his- ture; political economy
torical analysis of the platform’s boundary dynamics that ultimately
determine its operational scale and scope. Based on our analysis of
a unique set of archived primary sources, we discern four main
stages in Facebook’s long-term evolution and discuss the interplay
between ongoing processes of “platformisation” and “infra-
structuralisation”. We argue that these terms foreground comple-
mentary aspects of the platform’s efforts in balancing its expansion
and adaptability to changing user needs and other “environmental
dynamics” without risking its integrations and embedding in other
domains, such as advertising, marketing, and publishing. Ultimately,
our contribution illustrates how empirical platform histories can
denaturalise the current dominant position of social media plat-
forms, such as Facebook, revealing over a decade of incremental
evolution rather than revolution.

Introduction
The 2016 U.S. presidential elections marked the start of two years of negative coverage
of Facebook (Vaidhyanathan, 2018). Somewhat surprisingly, the company did not see
a decline in monthly active users, nor in its vast network of business partners. Indeed,
over the past decade, numerous companies worldwide have aligned their business
models and have integrated their technologies with Facebook. As of late 2018,
Facebook hosted over 90 million businesses and 6 million active advertisers (FIR-2018).
Despite collecting behavioural data of over 2.2 billion monthly active users, Facebook
hides a highly opaque, digital marketing ecosystem that keeps its revenue motor

CONTACT Anne Helmond [email protected] University of Amsterdam, Media Studies, Turfdraagsterpad 9,

1012 XT Amsterdam, the Netherlands.
ß 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives
License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in
any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
124 A. HELMOND ET AL.

humming. Collectively, these statistics point towards Facebook’s entrenchment: its

deep economic and infrastructural integration in the wider ecosystem of connective
media (van Dijck, 2013) and, in particular, the digital advertising and marketing indus-
tries (Crain, 2019). This entrenchment raises the question of how to account for the
platform’s economic growth and technological expansion as it caters to multiple stake-
holder groups (Helmond, Nieborg, & van der Vlist, 2017; Nieborg & Helmond, 2019).
In this article, we engage with Facebook’s evolution as a platform company that is
inherently unstable and subject to continuous change. We contribute to both popular
and academic efforts to historicise the company’s meteoric rise (Bru €gger, 2015;
Goggin, 2014; Kirkpatrick, 2010; Vaidhyanathan, 2018) by analysing Facebook’s ability
to forge and dissolve corporate partnerships over time. We contend that to under-
stand Facebook’s growth, it should be seen as not only a social networking site but
also a constantly changing platform that derives power from its ability to create insti-
tutional dependencies among its vast network of partners. It is easy to overlook this
techno-economic dimension of platform power. Early popular and scholarly contribu-
tions on “Web 2.0” emphasised the supposed emancipatory and democratic potential
of social media for users (van Dijck & Nieborg, 2009). Facebook’s chief executive Mark
Zuckerberg has been eager to maintain and broaden this framing by ceaselessly
“describing users as empowered social and political actors” (Hoffmann, Proferes &
Zimmer, 2018, p. 210). In this article, we focus on the process of Facebook positioning
itself not to its billions of end-users but rather to the very large business organisations
with which Facebook has partnered and that have been key drivers behind the plat-
form’s increased infrastructural power and presence.
Our empirical historical approach involves two complementary lines of enquiry.
First, we consider the dynamics of Facebook’s platform architecture and how its pro-
grammability evolved over the years to accommodate different stakeholder groups.
We investigate platform programmability by analysing Facebook’s evolving platform
boundaries, or more precisely, the platform-architectural borders that afford technical
and organisational interactions with external stakeholders, particularly corporate part-
ners. To that end, we draw on the notion of technical “boundary resources”
(Ghazawneh & Henfridsson, 2013), which include application programming interfaces
(APIs), software development kits (SDKs), and reference documentation. These resour-
ces enable external developers and organisations to build applications and integra-
tions with Facebook’s data and extend its functionalities. Second, we consider another
set of boundary resources: partnership programmes and related partner badges and
certifications. Ultimately, a platform’s control over this heterogeneous collection of
resources raises significant political-economic questions. When platforms are in full
control over (access to) their data infrastructure, individual platform-partnership rela-
tionships are inherently uneven and asymmetrical (Bechmann, 2013). In addition,
when pursuing technological expansion and economic growth, platforms deploy part-
nership strategies to connect and integrate with organisations worldwide that are
leading in other markets and industries. As a result, partnerships are an essential entry
point for tracing a platform’s evolution and its shifting boundaries.
These two complementary levels of analysis offer insights into the multiple forms of
boundary-work that platforms and partners perform together to drive their expansion
 INTERNET HISTORIES 125

and subsequent entrenchment, thereby gaining infrastructural properties and power.

On the one hand, platforms conduct internal boundary-work when negotiating their
“programmability” towards specific stakeholder groups (Mackenzie, 2018), which is
reflected in the formalisation and stabilisation of their platform architectures over
time. On the other hand, platforms conduct boundary-work with their partners. These
tend to be market leaders who are strategically positioned in other domains, such as
advertising, marketing, and publishing. Through these strategically forged corporate
partnerships, platforms instil platform dependencies, become embedded, and gain
power in other domains. Thus, we consider both internal and external boundary-work
as the operationalisation of infrastructural power, both of which are necessary to
establish connections with external partners.
By documenting Facebook’s shifting boundaries, especially in terms of its program-
mability and partnerships, we detail the platform’s constant, incremental reconfigur-
ation, which culminated in an aggressive infrastructural posture. In this regard, the
interplay between the processes of “platformisation” and “infrastructuralisation” cap-
tures how platforms expand their boundaries while simultaneously embedding them-
selves into other markets by strategically orienting their programmability towards
developers and businesses (Constantinides, Henfridsson, & Parker, 2018; Gerlitz &
Helmond, 2013; Helmond, 2015; Mackenzie, 2018; Plantin, et al. 2018). Consequently,
we trace Facebook’s transformation from a social network into a “platform-as-infra-
structure” (Plantin et al., 2018, p. 307).
Based on our publicly available unique data set of archived platform boundary
resources, we discern four stages in Facebook’s long-term evolutionary dynamics and
present these in the form of a periodisation, beginning in 2006. Our evolutionary per-
spective and methodological approach afford an empirical historical contribution to
platform studies that is deeply sensitive to the interplay between platforms and their
environments. In addition, we trace the incremental, minute modifications that, taken
together, point towards platform evolution rather than platform revolution.

Platform evolution and internet history

Similar to websites that are subject to “fluctuation” as they are editable and reprodu-
cible (Bru€gger & Finnemann, 2013), platforms change continuously. Typically, webmas-
ters and platform operators do not systematically document changes, nor do they
offer comprehensive archives of old materials, thereby posing serious challenges to
writing internet histories. Consequently, a key challenge for platform historians is “to
find useful sources that enable them to understand the evolutionary processes in the
first place” (Bruns & Weller, 2016, p. 186). In media and communications studies, mul-
tiple approaches have been employed to write platform histories (Bru €gger, 2015;
Burgess & Green, 2018; Elmer, 2017; Gerlitz & Helmond, 2013; Goggin, 2014; Hoffmann
et al., 2018; van Dijck, 2013; Rogers, 2013a, b). These approaches typically focus on the
evolution of a single platform and employ secondary sources, such as industry blog
posts and screenshots, to chronicle its evolution. However, as others have argued,
platforms evolve via a complex interplay among users, technologies, infrastructures,
126 A. HELMOND ET AL.

organisational structures, and various social, cultural, and economic practices (van
Dijck, 2013).
Our study differs from existing historical platform studies in four ways. First, we
move from single platform histories to an ecosystem-level view, which considers the
larger environments within which platforms operate. Second, our historical analysis
draws from a unique set of primary historical sources: archived platform boundary
resources made accessible by the Internet Archive Wayback Machine (cf. Helmond
et al., 2017; Nieborg & Helmond, 2019). Third, our approach mimics the foci and lan-
guage in the fields of information systems and organisation studies. In these fields,
which are typically not part of platform studies, “platform evolution” is studied con-
jointly with the evolution of digital infrastructures and inter-organisational networks
(Constantinides et al., 2018; de Reuver, Sørensen, & Basole, 2018). Similar to media and
communication scholars, organisational scholars have adopted biological models and
metaphors to conceptualise the dynamics of organisational structures (Mars &
Bronstein, 2018). For example, as digital platforms transform, their architectures, inte-
grations with partners, governance frameworks, and environmental contexts coevolve.
Collectively, these dynamics determine platforms’ “evolutionary trajectories”, particu-
larly in terms of “composability” and “malleability”, which are the two key short-term
evolutionary dynamics in a platform’s programmability (Tiwana, Konsynski, & Bush,
2010). As we detail, these two features describe the incremental changes of a plat-
form’s programmability and the ability of external developers and corporate partners
to extend platform functionality without compromising the platform’s integration
within the larger platform ecosystem. In short, they capture a platform’s technical
adaptation to changing user needs, technological innovation, market competition, and
other “environmental dynamics” (Tiwana et al., 2010).
Building on the notion of evolutionary trajectories, the fourth way we deviate from
historical platform studies is our level of temporal granularity. In our analysis, we dis-
tinguish between long-term and short-term evolutionary dynamics. Most historical
platform scholarship focuses on the former, offering broad-stroke histories based on
key events and leadership decisions impacting a platform’s design and governance.
These developments cover annual or multi-year periods. We complement such
accounts by including short-term dynamics that take place on a monthly or quarterly
basis. These include incremental, minute modifications in platform architecture or
boundary resources that ultimately underpin long-term shifts (e.g. achieving corporate
entrenchment, envelopment of competing platforms, derivative mutations such as dat-
ing or messaging platforms). Taken together, these dynamic adaptations are critical for
understanding the evolving programmability of platforms insofar as they can facilitate
external stakeholder groups, such as advertisers and publishers. In particular, these tra-
jectories reveal how platforms, through technical and partner-oriented resources, gov-
ern and control platform boundaries.

Platform boundaries and platform extensions

Platform evolution, we argue, can only be observed over a sufficiently long period of
time, which in the fast-paced platform economy means years, not decades. Therefore,
 INTERNET HISTORIES 127

in our historical approach, we suggest including a platform’s archived development

resources to allow for a longitudinal investigation of changes in a platform’s program-
mability. We build on the work by media studies scholars who examine the underlying
mechanisms and logics that structure a platform’s extensions into other domains and
markets. Examples of such extensions include software plug-ins, “social buttons”, and
other API-based connections that expand platform boundaries by integrating platform
data and features into third-party websites, software, and apps (Gerlitz & Helmond,
2013; Helmond, 2015).
Platform boundary resources, such as APIs, are important mechanisms to realise
platform extensions as they provide “a set of interfaces” that enables external web-
sites, platforms, and apps “to communicate, interact, and interoperate with the
platform” (Tiwana, 2014, p. 6). Consequently, they allow third-party developers, such
as marketing agencies, to build “on top of” a platform’s core infrastructure, thereby
extending its functionality. Relatedly, SDKs are important boundary resources that
facilitate and streamline the app development process by providing developers with a
set of software tools, developer libraries, APIs, documentation, code samples,
and guides.
Platform boundary resources are an important way for platform companies to exer-
cise power over their institutional relationships with third parties. On the one hand,
platforms change continuously and evolve alongside external contributors who inte-
grate platform functionality into their own software tools and products. On the other
hand, there is an incentive for platforms to maintain stability and standardise their
boundary resources for third-party development (Tiwana et al., 2010). The ability to
define platform architecture and governance is indicative of an “economy of data
intraoperability”, in which platform operators enforce asymmetrical institutional rela-
tionships with their partners (Bechmann, 2013). Tracing the evolution of Facebook’s
boundary resources helps us grasp not only how a platform’s architecture changes
and how its functionality becomes embedded in other domains, but also the evolution
of institutional dependency among corporate partners.

Corporate partnerships and exponential growth

Compared to more traditional companies in the information and communication
industries, one of the defining economic and organisational properties of platform
companies is their programmability. That is, they operate “multi-sided markets” by
bringing together different “sides” or “users” (Gawer, 2014; Tiwana, 2014). In the con-
text of multi-sided markets, users can be end-users (i.e. individuals) and a wide variety
of organisations, including but not limited to non-profits, governments, businesses,
content developers, and advertisers. A platform’s ability to thrive within an ecosystem
hinges on its ability to aggregate users (i.e. market sides) and facilitate seamless inter-
actions and transactions among them.
Most of the multi-sided market research is rooted in the fields of economics and
business studies (i.e. information technology and strategic management), which theor-
ise how firms can gain a competitive advantage by leveraging the externalities associ-
ated with networked markets (de Reuver et al., 2018; McIntyre & Srinivasan, 2017).
128 A. HELMOND ET AL.

Network externalities or “effects” describe how users accrue (or lose) value by other
users joining (or leaving) a platform. From an economic perspective, platform busi-
nesses are able to grow exponentially if they can grow all sides in the market as this
leads to cross-side externalities. For example, the more end-users join a market, the
more plentiful and valuable the transactions become for other sides in the market.
From a strategic management perspective, a platform’s “competitive advantage”
hinges on its ability to entice users to join a platform. Growing the pool of end-users
is typically an issue of scale: the bigger the pool, the higher the demand. Conversely,
growing organisational sides introduces supply-side economies of scope: heteroge-
neous organisations that partner with platforms not only offer products or services to
end-users, but also are positioned as “collaborative innovators” (Gawer, 2014, p. 1243).
In this role, they can introduce a wider variety of platform functionalities and extend a
platform’s core features. In the case of Facebook, this means that corporate partners,
such as advertising and marketing companies, can contribute technology, data, or
services that complement Facebook’s own products and services.
The scholarship on corporate partnerships is closely related to questions about plat-
form evolution and platform boundaries. Drawing from the fields of industrial organ-
isation and information systems, scholars studying multi-sided markets emphasise the
dynamic nature of platform design and how partners and technology are managed.
They argue that platform operators are incentivised to facilitate organisational align-
ment and integration among the various sides of a platform. As we noted earlier, plat-
form operators can accommodate corporate partners by offering a standardised,
stable, core technology (Tiwana, 2014). The fact that this is not always the case dem-
onstrates that the process of forging and sustaining organisational relationships is
fraught with tension, risk, and uncertainty. Because of the inherent power asymmetry
in platform ecosystems and the unbridled growth driven by network effects, the emer-
gence of “platform capitalism” has drawn the attention of critical political economists
(Bechmann, 2013; Nieborg & Poell, 2018; Srnicek, 2016). We align ourselves with these
critical perspectives and concur that corporate partnerships are inevitably entwined
with questions of power. Every additional corporate partnership solidifies a platform’s
infrastructural position and is one step closer to a more dominant position not only in
the platform ecosystem but also in broad and far-reaching markets and industries.

Tracing platform boundaries

To study Facebook’s evolving programmability and platform boundaries, we devel-
oped a methodological approach that uses archived platform materials to reconstruct
platform history. We retrieved these materials from the Internet Archive Wayback
Machine, which is the largest publicly-accessible web archive, containing over 344 bil-
lion “snapshots” from archived web pages since 1996. We also draw on Facebook’s
blog archives and trade publications (Appendix). Our data set begins in 2006 with the
launch of the Facebook Development Platform and ends in November 2018. In our
analysis, we partitioned our data into 14 intervals (avg. ¼ 1.4 intervals/year) to compare
different materials and moments. These intervals are based on prior exploratory
research (Helmond et al., 2017) that offered insights into key moments when changes
 INTERNET HISTORIES 129

occurred in our platform materials (i.e. developer, business, partner materials).

Additionally, to contextualise our historical analysis based on archived platform materi-
als, we conducted semi-structured, 30–60-minute background interviews with a small
number of Facebook’s marketing partners from 2013 to 2016. We interviewed found-
ers and business development executives of leading partner organisations, such as
App Annie, AppsFlyer, Fiksu, Grow Mobile, and TUNE, on-site or at industry events. We
draw on these interview materials to contextualise our analysis by including partners’
perception of the role and dynamics of Facebook’s partnerships. Finally, we used infor-
mation visualisations to present our historical reconstructions and the outcomes of
our analysis. All figures use identical temporal axes and intervals to enable compari-
son. Monochrome gradient bars are used to demarcate the four stages of our period-
isation, which we discuss after the analysis.
Our empirical analysis proceeds along two main lines of enquiry. First, we systemat-
ically retrieved archived snapshots of Facebook’s developer materials to trace the evo-
lution of Facebook’s programmability and relationships with different kinds of
developers. We then reconstructed changes in Facebook’s boundary resources, their
retrospective versioning, and the conditions under which third-party development and
external relations evolved. Second, we collected archived business and partner materi-
als to trace the evolution of Facebook’s relationships with partner organisations. Using
archived snapshots of partner programme directories which list all partnerships, we
took stock of all partners’ names and details. Partner programmes signal integrations
with officially approved or certified partner organisations who provide services or
implement platform data or products that augment Facebook’s reach and scale. We
then characterised these partnerships by examining their official partner badges and
specialties, which have a longer history online and typically function to mark
“authority, expertise, experience, and identity” (Halavais, 2012, pp. 356–357). In the
case of Facebook, these badges detail the particular capabilities and expertise by
which partners complement the platform (FB-2018a). We employed these materials to
trace the changing composition of Facebook’s ecosystem of marketing partners who
have been adjudicated on their “demonstrable expertise” and capacity to develop
apps that complement and extend Facebook’s own tools, products, and services.
In short, these materials allow us to reconstruct Facebook’s embeddedness within
larger technological, economic, and organisational structures. Our materials point to
Facebook’s ability to leverage cross-side network effects to expand the scale, scope,
and reach of its technical and business operations through strategic corporate partner-
ships. Neither boundary resources nor partner programmes are typically included in
historical platform studies.

Development of Facebook platform

In our first line of enquiry, we traced the evolution of Facebook’s programmability and
platform boundaries and visualised this over time (Figure 1). Launched in 2004 as a
social networking site, Facebook became programmable when it started inviting third-
party developers to integrate with Facebook: first with the beta launch of “Facebook
Development Platform” (2006), followed by “Facebook Platform” (2007) (FNo-2006;
130 A. HELMOND ET AL.

Figure 1. Evolution of Facebook’s platform, 2006–2018. https://doi.org/10.17605/osf.io/47zyc.

FNe-2007). To facilitate third-party development, Facebook offered a set of boundary

resources, which exposed the platform’s architecture and offered developers guidance
on how to access platform data and functionality to build their own applications.
In these formative years, Facebook Platform primarily focused on having third-party
developers build “social apps” inside Facebook’s domain and on top of its “social
graph”, which represents “the network of connections and relationships between peo-
ple” in the Facebook API (FNe-2007). At the F8 Developer Conference in 2010,
Facebook announced the first major iteration of their developer platform, dubbed
“v1.0”, which featured a new “Graph API”, formerly known as the “Facebook API.”
Since then, the platform has employed API versioning and so-called versioning sched-
ules to introduce regular updates and mark the deprecation of previous API versions
(FD-2018e). The launches coincided with the introduction of several SDKs to help
developers build mobile apps for Facebook, signalling its ambitions to expand into
the emerging mobile ecosystem (FD-2010b).
In mid-2010, the development platform was firmly in place, which paved the
way for the platform’s orientation towards businesses and advertising technology
companies. The introduction of Facebook’s “Ads API” meant that developers could
build their own advertising technologies on top of Facebook’s programmable plat-
form. The Ads API was available to selected “tools vendors” and marketing agen-
cies to create and manage their “ads on Facebook programmatically” (FD-2010a). It
offered partners deeper levels of technology integration by enabling them to con-
nect their own tools with Facebook’s advertising products, allowing partners to
 INTERNET HISTORIES 131

automate and manage ads on Facebook. As such, the rollout of the Ads API dem-
onstrates an important transition and expansion of Facebook’s development plat-
form by accommodating advertisers not only as customers, but also as a new
group of development partners.
In 2013–2015, flush with momentum and capital from its initial public offering (IPO)
in 2012, Facebook made a number of high-profile acquisitions to expand its user base
and advertising development platform. This is reflected by the increasing number of
boundary resources and the growing pace of API updates. In 2013 and 2014,
Facebook acquired Atlas, a programmatic advertising platform, and LiveRail, a video
advertising platform (FNe-2013; FNe-2014). Although both services were eventually dis-
continued, certain aspects of these platforms, such as the Atlas API, were integrated
into Facebook’s core advertising platform. Furthermore, Facebook expanded its focus
on mobile advertising by launching “Facebook Audience Network” (2014), which
included a set of boundary resources that enabled “advertisers to extend the scale of
their Facebook campaigns beyond Facebook and into other mobile apps” (FB-2014),
allowing advertisers to find and target audiences beyond the platform’s boundaries.
These acquisitions and the subsequent integration of external boundary resources
indicate how Facebook followed broader developments in digital marketing as the
company oriented itself towards programmatic advertising, video, and mobile advertis-
ing (Crain, 2019).
In 2015, Facebook officially rebranded the Ads API into the “Marketing API” (MAPI),
which can be seen as an effort to further broaden the scope of Facebook’s advertising
ambitions by explicitly hailing it as a platform for marketing development. In this con-
text, marketing refers to a broader set of corporate activities centred on promoting
and selling services, and typically includes market research and advertising. Together
with Facebook’s Audience Network for mobile advertising, foregrounding the MAPI’s
development marked a key moment in Facebook’s evolving programmability as it
enabled the development and integration of marketing apps. Finally, in 2018,
Facebook again redesigned and consolidated its technical boundary resources for busi-
nesses and marketing developers by integrating the platform resources of two of its
most popular apps, Instagram and Messenger, into its core platform.

Accrual of corporate partnerships

In our second line of enquiry, we examined how Facebook followed a multi-sided mar-
ket strategy and accrued corporate partnerships through 1) partner programmes and
2) their certification mechanisms (i.e. specialties and badges). This analysis set us up
for a third step: tracing the changing composition of Facebook’s marketing partner
ecosystem. While APIs and reference documentation are primarily aimed at developers,
these programmes and certifications are aimed at businesses and partners who build
the integrations that connect Facebook with adjacent markets and industries. We
reconstructed the evolution of Facebook’s partner programmes since 2007 and
observed a shift in orientation from partnering with developers, and then to advertis-
ing and marketing developers to media and content partners more broadly (Figure 2).
132 A. HELMOND ET AL.

Partner programmes
One of the earliest partner programmes, fbFund (2007–2009), awarded grants to
developers to build “their businesses on Facebook Platform” with “innovative and
engaging” apps (FD-2007). Additionally, the “Application Verification Program” and
“Great Apps Program” (2008–2009) were launched to create a “robust” and “thriving”
app ecosystem, pushing partners to build “meaningful”, “trustworthy”, and “well-
designed” apps. In return, verified app developers would obtain deeper platform inte-
grations, early access to new features, and support from Facebook’s growing partner
management team (FD-2008). The subsequent “Preferred Developer Consultant” (PDC)
programme (2009–2012) was aimed at connecting businesses with development part-
ners who were experienced in using Facebook products and technologies and had “a
long track record of providing Facebook-centric services to large Fortune 500 busi-
nesses” (FD-2009).
The next set of Facebook’s programmes focused on building and expanding its
advertising and marketing partnerships. The “Ads API Tools Vendors” programme
(2009–2011), later renamed the “Marketing API Program” (2011–2012), listed third-
party tools that were built by selected partners on top of the Ads API. The programme
aimed at connecting partners with access to the Ads API to major companies and
agencies to create and manage large Facebook advertising campaigns via these third-
party partner tools (FD-2009). Later, these programmes merged with the Preferred
Developer Consultant programme into the “Preferred Marketing Developer” (PMD) pro-
gramme (2012–2015), which was intended to find developers with the ability to build
comprehensive “solutions to Facebook marketing and business operations” (FNo-2011;
FD-2012a) and to create a “community of best-in-class developers focused on making
social marketing easier and more effective” (FPMDC-2013). The successive “Facebook
Marketing Partners” (FMP) programme (2015–present), and related marketing

Figure 2. Evolution of Facebook’s partner programmes, 2007–2018. https://doi.org/10.17605/osf.

io/47zyc.
 INTERNET HISTORIES 133

programmes such as the “Instagram Partners” programme (2015–present) and “Atlas

Partners” programme (2015–2017), further emphasised the development of marketing
technologies.
The Facebook Marketing Partner programme promised businesses to help find part-
ners who offer “innovative technology” and “custom-tailored solutions” to
“supercharge” their marketing efforts on and off Facebook (FMP-2015). The related
“Facebook Marketing API Accelerator Program” (2015) provided a “path to serious API
skills and support” to help marketing developers learn about the MAPI and Facebook
marketing (FD-2015). Such accelerator programmes, including fbFund (2007–2009) and
FbStart Partners (2014–present), provide developers with technical, educational, and
financial support to stimulate and facilitate the development of Facebook-integrated
business apps and marketing solutions, thereby contributing to the platform’s expan-
sion and embedding it further in the digital marketing ecosystem. The recent redesign
of Facebook Marketing Partners’ directory into “Solutions Explorer” (2018) was accom-
panied by the introduction of the “Facebook Marketing Consultants” (FMC) pro-
gramme (FD-2018a). These consultants are not fully vetted partners but individuals
who help smaller advertisers with their on-demand advertising and marketing needs
which “aren’t always addressed by our traditional partner ecosystem” (FD-2018d).
The latest phase of Facebook’s partner programmes shows an orientation towards
media partners, including broadcasters, publishers, and content providers, with pro-
grammes such as “Facebook Media Solutions” (2015–present) (Rein & Venturini, 2018).
Additionally, there are general public-oriented partner programmes such as the “Data
Abuse Bounty Program” and “Third-Party Fact-Checking Program”, which emerged in
response to recent critiques of Facebook concerning Cambridge Analytica and the
spread of misinformation, together with a programme that foregrounds the company’s
renewed focus on community building with the “Facebook Community Leadership
Program” (2018–present). These public programmes signal another phase in the evolu-
tion of the platform and its relations and responsibilities to end-users and external
stakeholders. The multiple types of partner programmes illustrate how Facebook inter-
acts with various stakeholder groups and how the platform truly became a multi-sided
platform, connecting app developers, advertisers, marketers, content producers, media,
and local communities.

Marketing partners’ specialties, badges, and certifications

An important aspect of the marketing partner programmes’ structure is the use of cer-
tifications. Here, we reconstructed how partners’ specialties and official badges
evolved over the years to determine the role of partners in Facebook’s expansion
(Figure 3). Changes in specialties and badges indicate how and when Facebook shifted
its orientation from platform-centric advertising services to business solutions that are
familiar to a broader set of digital advertisers and marketers.
In 2010, Facebook offered a single Preferred Developer Consultant badge that part-
ners could put on their websites to indicate a sanctioned relationship with the plat-
form. Partners only had three “Areas of Expertise” (i.e. “Connect”, “Applications on
Facebook.com”, and “Pages”), which were intended to build “deeply integrated social
134 A. HELMOND ET AL.

Figure 3. Evolution of Facebook’s marketing partner specialties, official partner badges, and certifi-
cations, 2010–2018. https://doi.org/10.17605/osf.io/47zyc.

experiences” within the confines of the platform or across its boundaries with
“Connect” (now “Facebook Login”). In 2012, this list grew to an extensive list of
unstructured, self-defined areas of expertise. Newly minted “Preferred Marketing
Developers” (PMDs) received a new badge displaying up to four main “qualifications”:
“Ads”, “Apps”, “Insights”, and “Pages.” In the words of Facebook, certified partners
“extend measurably beyond the functionality of Facebook’s native tools” (FD-2012b).
Later, the programme created a special badge for “Strategic PMDs” for a select group
of “top marketing developers” who are “driving outstanding results in the Facebook
marketing developer ecosystem” and who, in return, receive the highest level of sup-
port (FS-2012).
In 2013–2014, several new “qualifications” were added to the Preferred Marketing
Developers programme for (i) “FBX Qualified Companies”, who successfully integrated
with Facebook’s programmatic advertising exchange, (ii) “Agencies with Ads API
Access”, who qualified for API access but not for an official partner badge, and (iii)
“Mobile Measurement Qualified Companies”, who provided tools for mobile ad cam-
paigns’ performance measurements. These new specialties can be seen as part of
Facebook’s “mobile career” (Goggin, 2014). One of our interviewees at a leading
mobile app tracking company states that “being a Facebook Mobile Measurement
partner helps us” and “puts us on a good standing to work with a lot of advertisers”
as they are one of a select few who are authorised to run and track campaigns on
Facebook. Nevertheless, this partner also voiced concerns over how this authorisation
can easily be retracted, stating that “Facebook always holds a lot of power” over its
partners (I-2015a). This type of platform power is further apparent in an interview with
an early Mobile Measurement partner that measured app installs for Facebook, who
was removed from the programme in 2014 for allegedly violating the platform’s terms
of services regarding data retention (I-2015b). According to the partner, Facebook
 INTERNET HISTORIES 135

“built enough value around their product that people need it and because of that
they set their own rules”. This partner concluded that platforms such as Facebook
“want to control the entire environment for app developers” (I-2015b). Thus, while
partnerships are generally considered mutually beneficial, platform–partner relations
are inherently asymmetric.
In 2015, the Facebook Marketing Partners programme introduced a single
“Marketing Partner” badge to represent multiple “Specialties”, which no longer referred
directly to platform-centric business products but instead employed common profes-
sional marketing terminology (e.g. “Ad Technology”, “Content Marketing”, “Media
Buying (US Only)”). This updated terminology indicates how Facebook seeks to inte-
grate the distinct tools, products, and services of its platforms into a single, unified
marketing platform accessible to partners, using general marketing terminology. In
this period, specialties such as “Audience Onboarding” and “Audience Data Providers”
arose to enable marketers to find existing customers on Facebook using a marketer’s
own data and to create new audience profiles on Facebook with the help of third-
party data partners. With the growth of its mobile app products, Instagram’s market-
ing partner specialties were aligned with Facebook’s, by employing the same special-
ties and badges. Notably, in early 2018, as a response to the Facebook–Cambridge
Analytica data scandal, the “Audience Data Providers” specialty was removed.
Finally, official partner badges also signal certifications in knowledge and learning
(cf. Halavais, 2012, p. 369), such as Facebook Blueprint’s “Certification Badges”
(2015–present). Blueprint is an “education program that trains agencies, partners and
marketers on how to use Facebook” to create “better campaigns” through online
courses and exams (FB-2015). However, it is not merely a training programme because
some Marketing Partner specialties require the completion of Blueprint courses
(FB-2018b).

Marketing partner ecosystem

In addition to examining partner programmes and certification mechanisms, we also
traced changes in the composition of Facebook’s marketing partner ecosystem. We
used archived partner materials to reconstruct the evolution and dynamics of partner-
ships in the partner ecosystem (Figure 4). In total, we found 3,129 partner names over
the period 2009–2018 (1,033 unique partners). The number of partners increased most
between 2009–2013, with the largest number of additions and removals between
2012–2013. Increases in partnerships correspond with major changes to the develop-
ment platform and newly launched partner programmes, suggesting that they are
used to attract developers, businesses, content producers, and publishers to adopt
newly launched platform tools and products. Between 2014–2018, the total number of
partners remained more constant, although in 2018 there are many new partners with
the introduction of the Facebook Marketing Consultants programme as part of the
Facebook Marketing Partners programme.
While there are partnership additions and removals at every interval, as many as 42
of our partners appear more than 10 times across our 14 intervals (avg. length ¼
7.1 years). These long-term partners are leading digital advertising and marketing
136 A. HELMOND ET AL.

Figure 4. Evolution of Facebook’s marketing partner ecosystem, 2009–2018. https://doi.org/10.

17605/osf.io/47zyc.

technology firms such as Brand Networks, Kenshoo, Nanigans, Adobe, SocialCode,

AdParlor, Adaptly, Marin Software, and Salesforce, all of which have integrated their
own platforms with Facebook. These kinds of partnerships indicate how Facebook is
entangled both computationally and organisationally with the global network of lead-
ing advertising and marketing technology companies. One of the mobile advertising
companies we interviewed has been a long-term partner since 2011 and is technically
integrated with over 130 distinct advertising networks and major traffic sources such
as Facebook. It describes itself as helping advertisers to “navigate through a really
messy ecosystem” of interconnected platforms, each of which performs a specific task.
For smaller firms such as these, of which there are many, a partnership with Facebook
is not only strategic but also deemed essential (I-2016).
By tracing the changing composition of Facebook’s partner ecosystem and describ-
ing partners’ categories, we gained insights into Facebook’s larger embedding in
digital advertising and marketing technologies, other markets, industries, and coun-
tries. We matched our entire list of partners to the annual marketing technology data
 INTERNET HISTORIES 137

set released on chiefmartec.com, a reputable market research blog since 2011 (Brinker,
2018). The 2018 data set lists 6,829 distinct marketing technology solutions and their
categories, which we employed to characterise Facebook’s partnerships and under-
stand their embedding in the marketing technology industry landscape (Figure 5).
Partners were mostly specialised in the categories of “Advertising & Promotion”
(598), “Social and Relationships” (375), “Data” (294), and “Content & Experience” (267),
while “Commerce & Sales” (95) and “Management” (36) did not have a significant pres-
ence. Notably, “Advertising & Promotion” and “Data” rose in prominence between
2012–2014, reflecting Facebook’s orientation towards advertising technology and its
growing prominence as a data platform within the industry. “Content & Experience”
has been steadily growing since 2009, pointing to Facebook’s key role in the platform-
isation of cultural production (cf. Nieborg & Poell, 2018).
On a sub-category level, we observed the rise of “Search & Social Advertising” (297)
and “Display & Programmatic Advertising” (205), especially between 2012–2016.
Display and programmatic partnerships declined since 2016 due to the shutdown of
Facebook Exchange (2015), its ad exchange. The growth of mobile-oriented partner-
ships (e.g. “Mobile & Web Analytics”, “Mobile Marketing”) reflects Facebook’s mobile
orientation since the mid-2010s (Goggin, 2014), as well as a larger industry-level shift
towards “mobile-first”. First in 2012, then in 2015, there was an increase in partners
engaging in “Social Media Marketing & Monitoring” (238), reflecting the popularity of
tools for online brand presence and community management on Facebook. Also, since
2012, Facebook has accrued many data-oriented partnerships in “Audience/Marketing
Data & Data Enhancement” (86) and “DMP” (60) – or Data Management Platforms,
which combine the collection, organisation, analysis, and activation of data for target-
ing and analytics purposes. We further found a long tail of more widely-oriented part-
nerships across all categories.

Periodisation of Facebook’s evolution

Surveying Facebook’s decade-long deployment of boundary resources provides the
basis for a periodisation of its evolution. We discern four stages that together charac-
terise key moments in Facebook’s programmability and expansion of its boundaries.
The purpose is not to discretise the historical developments as clear-cut periods but
rather to characterise some long-term developments in Facebook’s evolution and
thereby offer analytical handles for understanding the historical precedents of its
transformation into a “platform-as-infrastructure” (Plantin et al., 2018, p. 307). In our
periodisation, particular development efforts introduced in earlier stages are built
upon, extended, and integrated, or alternatively, discontinued and deprecated in sub-
sequent stages.
Stage one in Facebook’s evolution (2006–2010) concerns the expansion of its social
networking site, with the launch of the Facebook Development Platform. Facebook
started attracting third-party developers by offering boundary resources and financial
and technical support to accelerate “good” app development (e.g. fbFund and Great
Apps Program), thereby embedding itself into the developer community. The
Preferred Developer Consultant programme helped brands and businesses to grow a
138 A. HELMOND ET AL.

Figure 5. Evolution of Facebook’s embedding in the marketing technology landscape, 2009–2018.

Categories: 2018 Marketing Technology Landscape Supergraphic (Brinker, 2018). https://doi.org/10.
17605/osf.io/47zyc.

Facebook presence, build apps, and accommodate the enrolment of high-profile part-
ner organisations. Additionally, the Ads API and tools vendors programme were key
initiatives to explore and extend the programmability of Facebook’s platform towards
a new stakeholder group of advertising developers. Despite being only available to a
select few, these resources mark the early onset of Facebook’s advertising develop-
ment platform.
Stage two (2010–2014) surrounds Facebook’s IPO in May 2012. Already, we can
observe Facebook’s infrastructural ambitions based on the maturation of its advertis-
ing development platform alongside its development platform. In both cases,
Facebook’s embedding was achieved through the development of apps and integra-
tions. During this period, the Ads API morphed into the MAPI, which signalled an
 INTERNET HISTORIES 139

ambition to grow the business side of the platform beyond advertising to include
other marketing products and services such as programmatic advertising, analytics,
and insights. The accompanying partner programme enrolled partners capable of
implementing Facebook’s marketing products into their own software platforms,
thereby further expanding Facebook’s platform boundaries, its capabilities, and the
reach of its technical and business operations. Through engaging in strategic partner-
ships with leading firms, Facebook legitimised itself not only as a viable advertising
platform but also as a one-stop-shop marketing platform. This is also reflected in the
merging of several partner programmes into a single Preferred Marketing Developer
programme to accommodate and attract new marketing developers. Facebook’s part-
ners became vital in this effort by slotting themselves into Facebook-specific special-
ties conceived around its core platform-centric business products at the time (i.e. Ads,
Apps, Pages, Insights). Furthermore, by adopting official partner badges, these partner-
ships legitimised Facebook’s prominent position as a core player in digital advertising
and marketing.
Stage three (2014–2018) revolves around the solidification and continued profes-
sionalisation of Facebook’s marketing development platform and its integrations in
other global markets and industries. Facebook’s two main development platforms
adopted a “core and extended versioning model” with regular API releases and
scheduled deprecation dates (FD-2018e). These communicative standards enable the
growing developer and marketing developer communities to anticipate the main-
tenance work required to ensure their apps and integrations, upon which their
businesses increasingly depend, will continue to work. Additionally, Facebook made
a number of high-profile acquisitions, including Instagram (already in 2012),
WhatsApp (2014), Oculus VR (2014), and LiveRail (2014). Their acquired development
platforms and boundary resources were gradually streamlined into the Marketing
API and Facebook Marketing Partners programme. The MAPI Accelerator Program
provided developers with additional resources to work with Facebook’s APIs to
facilitate the platform’s integration in other markets and industries, which enabled
its technical and business operations to reach even further. Furthermore, Blueprint
was launched to offer marketers and agencies training and certifications for
Facebook’s marketing tools and products. This coincided with another round of
partner programmes by which Facebook addressed new stakeholder groups in
media and publishing, content production, and local (developer) communities. As
media and content partners gained visibility, Facebook further grew from a user-
generated content site into a site for professional content producers and
media publishers.
Stage four (2018–present) marks Facebook’s current efforts to address criticism
about its market dominance and shortcomings with new programmes to combat data
abuse and misinformation by offering new programmes and governance mechanisms
for Facebook’s boundary resources (FD-2018b). This is accompanied by a major
redesign and restructuring of Facebook’s developer pages, business pages, and partner
pages as part of Facebook’s larger effort to “reexamine our platform” for building end-
user and developer trust (FD-2018c). These changes occurred with the v3.0 release of
Facebook’s platform APIs, which fully incorporates all Facebook products, including
140 A. HELMOND ET AL.

the Instagram Graph API. This is also reflected in the new unified Solutions Explorer
with marketing partner programmes that cover Facebook’s “family of apps” –
Facebook, Instagram, WhatsApp, and Messenger – and services. Finally, the new
Facebook Marketing Consultants programme introduces individual consultants who
can establish Facebook marketing technologies for smaller advertisers and businesses
not addressed by the partner ecosystem.
These four periods summarise Facebook’s long-term evolutionary trajectory as
shaped by the complex interplay between its platform architecture and the dynamics
of its technical and organisational environment. We contend that tracing the evolution
of Facebook’s programmability and corporate partnerships is key to understanding
these dynamics and the gradual accumulation of influence and power through the
processes of platformisation and infrastructuralisation. On the one hand, the compos-
ability and malleability of Facebook’s platform architecture enable partners to deploy
Facebook’s data and functionalities with relative ease while simultaneously enabling
Facebook to govern and control the conditions under which these can be reconfig-
ured (cf. Tiwana et al., 2010). On the other hand, Facebook’s corporate partnerships,
particularly with market-leading, global firms, facilitate its rapid entry into new mar-
kets, thereby generating and solidifying asymmetrical platform growth and dependen-
cies. Although such developments are often conceived in terms of innovation and
disruption, they are in many ways better characterised as ongoing boundary-work
with incremental, short-term effects that may (or may not) result in long-term
transformations.

Conclusion: evolution of a platform-as-infrastructure

To advance historical platform studies, we operationalised an empirical evolutionary
perspective on platforms and platforms-as-infrastructure. We examined how social
media platforms, such as Facebook, evolve as programmable architectures and, via
integrations with corporate partners, as businesses. By drawing on a unique set of
primary historical sources, we offered a methodological approach to chronicle these
evolutionary trajectories. Facebook’s archived platform boundary resources enabled
us to trace the platform’s shifting boundaries and the boundary-work that underpin
its exponential growth and embedding in other domains, especially advertising and
marketing. In particular, we traced the articulation of Facebook’s platform bounda-
ries through two complementary lines of enquiry. On the one hand, we recon-
structed the evolution of its programmability as facilitated and governed by APIs,
SDKs, and related boundary resources. On the other hand, we reconstructed the
evolution of its corporate partnerships, which were especially important in develop-
ing the apps and integrations that connected Facebook with adjacent markets and
industries, thereby extending the platform’s power. As such, we contribute to plat-
form studies by providing an infrastructural perspective on Facebook’s growth by
highlighting the role of partnerships. Yet, although many of Facebook’s partnerships
are publicly listed, recent revelations suggest that there are also non-public partner-
ships with organisations that have been “whitelisted” for special API access (Collins,
 INTERNET HISTORIES 141

2018). Further research could determine the implications of these non-public

partnerships.
Our findings indicate that social networks such as Facebook were not infrastructural
at launch, but rather gained infrastructural properties over time by accumulating
external dependencies through computational and organisational platform integra-
tions. First, in terms of evolving platform boundaries, Facebook has been steadily
growing by accommodating various strategic stakeholder groups through its architec-
tural design and programmability. In particular, its programmability has facilitated mul-
tiple developer communities to embed Facebook’s platform and operations in various
other domains, including software development, advertising, marketing, content pro-
duction, and media publishing. Thus, the platform changed from a social networking
site into a multi-sided platform for “social” app development, advertising develop-
ment, and marketing development. It did so through internal boundary-work concern-
ing the programmability of its platforms and through cooperative boundary-work with
partner organisations, mediated through domain-specific developer communities.
Second, in terms of its evolving embedding, Facebook has accumulated external
dependencies by routing additional technical and business operations and stakeholder
interactions through its platform. As an organisation, Facebook moved from a stand-
alone technology company to a public holding company with stakeholders and share-
holders. Today, it operates a single unified data infrastructure that gives way to a
number of “platform instances”, such as Messenger and Instagram, each of which con-
tributes to the platform’s boundary-work (Nieborg & Helmond, 2019). Lastly, Facebook
developed from a small online advertising business into a leading data-driven advertis-
ing and marketing platform, as well as a content monetisation platform for creators
and publishers (Nieborg & Poell, 2018).
The interplay between the processes of platformisation and infrastructuralisation
thus foregrounds different aspects of Facebook’s economic growth and technological
expansion. While platformisation speaks to Facebook’s growing capabilities to mediate
the interactions between multiple stakeholder groups and their diverging needs and
interests, infrastructuralisation speaks to Facebook’s growing ubiquity by embedding
itself in other markets and industries to render technical and business operations
more widely and immediately available. Indeed, infrastructure is not simply an analyt-
ical concept; becoming infrastructural is an effective platform strategy to “survive in
the long run” (de Reuver et al., 2018). Thus, platform power is as much economic,
operationalised by access to finance capital (Elmer, 2017), as it is relational through
Facebook’s ability to mandate organisational alignment among its stakeholders.
Therefore, both processes highlight different aspects of the boundary-work that
Facebook and its partners perform, as well as the political-economic stakes and conse-
quences of such work. This cooperative boundary-work embeds the platform in other
domains and removes barriers to entry, while at the same time avoids sectoral liability
and responsibility (van Dijck, Poell, & de Waal, 2018). In particular, what we contribute
to platform studies is a way to analyse how both short- and long-term developments
constitute platform power. Furthermore, since most leading social media platforms fol-
low similar development trajectories – they also operate partner programmes, devel-
opment platforms, and advertising and marketing platforms whose materials have
142 A. HELMOND ET AL.

been archived – there are ample opportunities for comparative historical plat-
form research.
Finally, there is a critical need for additional historical platform and platform-as-infra-
structure research to denaturalise the present market dominance of platform companies
such as Facebook. Because power and influence are relational concepts, critical platform
histories should consider the platform not only as an ensemble of technical elements,
but also as the relational intersection of multiple stakeholders that are embedded in vari-
ous domains, regions, and markets. Although social media platforms, at first glance, pose
challenges for internet history due to their constant updates, their archived platform
materials afford new kinds of detailed, empirical histories. These materials can be used to
trace the short-term, minor, and incremental changes that platforms undergo, thereby
countering popular myths of ensuing radical innovation and platform revolution.

Acknowledgements
All authors contributed equally to this work. Earlier versions of this article were presented at the
2017 Data Power Conference (Ottawa, Canada, June 22–23, 2017), and the 8th International
Conference on Social Media & Society (Toronto, Canada, July 28–30, 2017), and published in the
#SMSociety Proceedings (Helmond et al., 2017). The authors express appreciation to the anonym-
ous reviewers, whose constructive feedback greatly improved the article.

Data availability
The data that support the findings of this study are openly available in Open Science
Framework (OSF) at https://doi.org/10.17605/osf.io/47zyc.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek
under Grant 275-45-009; and the Deutsche Forschungsgemeinschaft under Grant DFG-SFB-1187.

Notes on contributors
Anne Helmond is an assistant professor of New Media and Digital Culture at the University of
Amsterdam. Her research interests include software studies, platform studies, app studies, digital
methods, and web history.
David Nieborg is an assistant professor of Media Studies at the University of Toronto. His
research interests include the political economy of platforms, the transformation of the game
industry, and games journalism.
Fernando van der Vlist is a research associate with the Collaborative Research Centre “Media of
Cooperation” at the University of Siegen and a PhD candidate at Utrecht University. His research
interests include software studies, digital methods, social media and platform studies, app stud-
ies, and critical data studies.
 INTERNET HISTORIES 143

ORCID
Anne Helmond http://orcid.org/0000-0003-4327-4012
David B. Nieborg http://orcid.org/0000-0002-3498-789X
Fernando N. van der Vlist http://orcid.org/0000-0003-1401-0325

References
Bechmann, A. (2013). Internet profiling: The economy of data intraoperability on Facebook and
Google. MedieKultur: Journal of Media and Communication Research, 29(55), 72–91. doi:10.
7146/mediekultur.v29i55.8070.
Brinker, S. (2018, April 24). Marketing technology landscape supergraphic (2018): Martech 5000
(actually 6,829). Retrieved from https://chiefmartec.com/2018/04/marketing-technology-land-
scape-supergraphic-2018/.
Br€
ugger, N. (2015). A brief history of Facebook as a media text: The development of an empty
structure. First Monday, 20(5). doi:10.5210/fm.v20i5.5423.
Br€
ugger, N., & Finnemann, N. O. (2013). The web and digital humanities: Theoretical and meth-
odological concerns. Journal of Broadcasting & Electronic Media, 57(1), 66–80. doi:10.1080/
08838151.2012.761699
Bruns, A., & Weller, K. (2016). Twitter as a first draft of the present: And the challenges of pre-
serving it for the future. In WebSci’16 Proceedings (pp. 183–189). New York, NY: ACM. https://
doi.org/10.1145/2908131.2908174
Burgess, J., & Green, J. (2018). YouTube: Online Video and Participatory Culture (2nd edn).
Cambridge, UK: Polity.
Collins, D. N. T. (2018, December 5) Note by Damian Collins MP, Chair of the DCMS Committee,
and selected documents ordered from Six4Three. Retrieved from: https://www.parliament.uk/
documents/commons-committees/culture-media-and-sport/Note-by-Chair-and-selected-docum
ents-ordered-from-Six4Three.pdf.
Constantinides, P., Henfridsson, O., & Parker, G. G. (2018). Introduction—platforms and infrastruc-
tures in the digital age. Information Systems Research, 29(2), 381–400. doi:10.1287/
isre.2018.0794
Crain, M. (2019). A critical political economy of web advertising history. In N. Br€ ugger & I.
Milligan (Eds.), The SAGE Handbook of Web History. London, UK: SAGE.
Elmer, G. (2017). Precorporation: Or what financialisation can tell us about the histories of the
Internet. Internet Histories, 1(1-2), 90–96. doi:10.1080/24701475.2017.1308197
Gawer, A. (2014). Bridging differing perspectives on technological platforms: Toward an integra-
tive framework. Research Policy, 43(7), 1239–1249. doi:10.1016/j.respol.2014.03.006
Gerlitz, C., & Helmond, A. (2013). The like economy: Social buttons and the data-intensive web.
New Media & Society, 15(8), 1348–1365. doi:10.1177/1461444812472322
Ghazawneh, A., & Henfridsson, O. (2013). Balancing platform control and external contribution in
third-party development: The boundary resources model. Information Systems Journal, 23(2),
173–192. doi:10.1111/j.1365-2575.2012.00406.x
Goggin, G. (2014). Facebook’s mobile career. New Media & Society, 16(7), 1068–1086. doi:
10.1177/1461444814543996
Halavais, A. M. C. (2012). A genealogy of badges. Information, Communication & Society, 15(3),
354–373. doi:10.1080/1369118X.2011.641992
Helmond, A. (2015). The platformization of the web: Making web data platform ready. Social
Media þ Society, 1(2). doi:10.1177/2056305115603080.
Helmond, A., Nieborg, D. B., & van der Vlist, F. N. (2017). The political economy of social data: A
historical analysis of platform-industry partnerships. In #SMSociety’17 Proceedings (pp. 1–5).
New York, NY: ACM. https://doi.org/10.1145/3097286.3097324
144 A. HELMOND ET AL.

Hoffmann, A. L., Proferes, N., & Zimmer, M. (2018). “Making the world more open and con-
nected”: Mark Zuckerberg and the discursive construction of Facebook and its users. New
Media & Society, 20(1), 199–218. doi:10.1177/1461444816660784
Kirkpatrick, D. (2010). The Facebook effect: The real inside story of Mark Zuckerberg and the world’s
fastest growing company. New York, NY: Random House.
Mackenzie, A. (2018). From API to AI: platforms and their opacities. Information, Communication
& Society., doi:10.1080/1369118X.2018.1476569
Mars, M. M., & Bronstein, J. L. (2018). The promise of the organizational ecosystem metaphor: An
argument for biological rigor. Journal of Management Inquiry, 27(4), 382–391. doi:10.1177/
1056492617706546
McIntyre, D. P., & Srinivasan, A. (2017). Networks, platforms, and strategy: Emerging views and
next steps. Strategic Management Journal, 38(1), 141–160. doi:10.1002/smj.2596
Nieborg, D. B., & Helmond, A. (2019). The political economy of Facebook’s platformization in the
mobile ecosystem: Facebook Messenger as a platform instance. Media, Culture & Society, 41(2),
196–218. doi:10.1177/0163443718818384
Nieborg, D. B., & Poell, T. (2018). The platformization of cultural production: Theorizing the con-
tingent cultural commodity. New Media & Society, 20(11), 4275–4292. doi:10.1177/
1461444818769694
Plantin, J.-C., Lagoze, C., Edwards, P. N., & Sandvig, C. (2018). Infrastructure studies meet platform
studies in the age of Google and Facebook. New Media & Society, 20(1), 293–310. doi:10.1177/
1461444816661553
Rein, K., & Venturini, T. (2018). Ploughing digital landscapes: How Facebook influences the evolu-
tion of live video streaming. New Media & Society, 20(9), 3359–3380. doi:10.1177/
1461444817748954
de Reuver, M., Sørensen, C., & Basole, R. C. (2018). The digital platform: A research agenda.
Journal of Information Technology, 33(2), 124–135. doi:10.1057/s41265-016-0033-3
Rogers, R. (2013a). Debanalizing Twitter: The transformation of an object of study. In WebSci’13
Proceedings (pp. 356–365). New York, NY: ACM. https://doi.org/10.1145/2464464.2464511.
Rogers, R. (2013b). Digital methods. Cambridge, MA: MIT Press.
Srnicek, N. (2016). Platform capitalism. Cambridge, UK : Polity.
Tiwana, A. (2014). Platform ecosystems: Aligning architecture, governance, and strategy. Waltham,
MA: Kaufmann.
Tiwana, A., Konsynski, B., & Bush, A. A. (2010). Research commentary—Platform evolution:
Coevolution of platform architecture, governance, and environmental dynamics. Information
Systems Research, 21(4), 675–687. doi:10.1287/isre.1100.0323
Vaidhyanathan, S. (2018). Antisocial media: How Facebook disconnects us and undermines democ-
racy. New York, NY: Oxford University Press.
van Dijck, J. (2013). The culture of connectivity: A critical history of social media. New York, NY:
Oxford University Press.
van Dijck, J., & Nieborg, D. B. (2009). Wikinomics and its discontents: A critical analysis of Web
2.0 business manifestos. New Media & Society, 11(5), 855–874. doi:10.1177/1461444809105356
van Dijck, J., Poell, T., & de Waal, M. (2018). The Platform Society: Public Values in a Connective
World. New York, NY: Oxford University Press.
 INTERNET HISTORIES 145

Appendix

Table A-1. References to original sources, sorted by type and date.

In-text citation Type Reference
FB-2014 Business (2014, October 7). The Audience Network is now available to more market-
ers worldwide. Retrieved from https://www.facebook.com/business/
news/audience-network-launch
FB-2015 Business (2015, March 24). New resources for Facebook marketers. Retrieved from
https://www.facebook.com/business/news/education-resources-
for-marketers
FB-2018a Business (2018a, November 8). Become a Facebook Marketing Partner. Retrieved
from https://www.facebook.com/business/marketing-partners/become-
a-partner
FB-2018b Business (2018b, November 8). Facebook Ad Technology Specialty Requirements.
Retrieved from https://www.facebook.com/business/marketing-partners/
adtech-speciality-requirements
FD-2007 Developers Vora, A. (2007, October 15). fbFund info and submission process. Retrieved
from https://web.archive.org/web/20090618172234/http://developers.
facebook.com/news.php?blog ¼ 1&story ¼ 38
FD-2008 Developers Ling, B. (2008, July 25). New ways for us to partner together. Retrieved
from https://web.archive.org/web/20090618161116/http://developers.
facebook.com/news.php?blog ¼ 1&story ¼ 144
FD-2009 Developers Thayer, K. (2009, December 16). Introducing the Preferred Developer
Consultant program. Retrieved from https://developers.facebook.com/
blog/post/346/
FD-2010a Developers (2010a, December 5). Ads API. Retrieved from http://web.archive.org/web/
20101205131945/http://developers.facebook.com/docs/adsapi
FD-2010b Developers (2010b, December 5). Mobile applications. Retrieved from https://web.arch-
ive.org/web/20101205131421/http://developers.facebook.com/docs/
guides/mobile/
FD-2012a Developers Yi, J. (2012a, April 18). The New Preferred Marketing Developer Program.
Retrieved from https://developers.facebook.com/blog/post/2012/04/18/
the-new-preferred-marketing-developer-program/
FD-2012b Developers (2012b, May 1). How to get the badge. Retrieved from https://web.archive.
org/web/20120501084120/https://developers.facebook.com/preferred-
marketingdevelopers/get_badge/
FD-2015 Developers Lu, C. (2015, November 11). Announcing the Marketing API Accelerator
and Virtual Hack. Retrieved from https://developers.facebook.com/ads/
blog/post/2015/11/11/announcing-accelerator-and-virtual-hack/
FD-2018a Developers Ho, C. (2018a, March 5). New directory tool for Facebook Marketing
Partners. Retrieved from https://developers.facebook.com/ads/blog/post/
2018/03/05/solutions-explorer-directory/
FD-2018b Developers Archibong, I. (2018b, April 4). API and Other Platform Product Changes.
Retrieved from https://developers.facebook.com/blog/post/2018/04/04/
facebook-api-platform-product-changes/
FD-2018c Developers Archibong, I. (2018c, May 1). Investing in the Facebook Developer
Community. Retrieved from https://developers.facebook.com/blog/post/
2018/05/01/investing-in-facebook-developer-community/
FD-2018d Developers White, S. (2018d, October 29). Introducing the Facebook Marketing
Consultants program. Retrieved from https://developers.facebook.com/
ads/blog/post/2018/10/29/introducing-fmc-program/
FD-2018e Developers (2018e, November 8). Versioning - Marketing API - Documentation.
Retrieved from https://developers.facebook.com/docs/marketing-
api/versions/
FIR-2018 Investor Relations (2018, October 30). Facebook Q3 2018 earnings. Retrieved from https://
investor.fb.com/investor-events/event-details/2018/Facebook-Q3-2018-
Earnings/default.aspx
FMP-2015 Marketing Partners (2015, February 22). Facebook Marketing Partners. Retrieved from https://
web.archive.org/web/20150222032351/http://facebookmarketingpart-
ners.com/
(continued)
146 A. HELMOND ET AL.

Table A-1. Continued.

In-text citation Type Reference
FNe-2007 Newsroom (2007, May 24). Facebook Platform launches with 65 developer partners
and over 85 applications for Facebook. Retrieved from https://news-
room.fb.com/news/2007/05/facebook-platform-launches-with-65-devel-
oper-partners-and-over-85-applications-for-facebook/
FNe-2013 Newsroom Boland, B. (2013, February 28). Facebook to acquire Atlas from Microsoft.
Retrieved from https://newsroom.fb.com/news/2013/02/facebook-to-
acquire-atlas-from-microsoft/
FNe-2014 Newsroom Boland, B. (2014, July 2). Facebook to acquire LiveRail. Retrieved from
https://newsroom.fb.com/news/2014/07/facebook-to-acquire-liverail/
FNo-2006 Notes Fetterman, D. (2006, August 15). Facebook Development Platform
launches … . Retrieved from https://www.facebook.com/
notes/2207512130
FNo-2011 Notes (2011, November 14). Preferred Developer Consultant Program. Retrieved
from https://www.facebook.com/notes/facebook/preferred-developer-
consultant-program/281113178594498/
FPMDC-2013 PMD Center (2013, January 18). PMD Center. Retrieved from http://web.archive.org/
web/20130118122604/http://www.facebook-pmdcenter.com:80/
FS-2012 Studio (2012, September 21). Announcing a new Designation for top marketing
developers. Retrieved from https://web.archive.org/web/
20120923004937/http://www.facebook-studio.com/news/item/announc-
ing-a-new-designation-for-top-marketing-developers
I-2015a Interviews (2015, February 4). Mobile Measurement Qualified Company.
I-2015b Interviews (2015, February 4). Mobile Measurement Qualified Company.
I-2016 Interviews (2016, February 13). Facebook Marketing Partner.