Business models innovation based on crowds: a comparative study

Over the centuries, technological progress changed the way businesses conducted their activities. However, the arrival of information technology (IT) has been one of the most powerful drivers of business model evolution. The advent of IT irreversibly changed the way businesses function, for example, the way they operate their supply chains, store information, contact their customers, and so on.

Businesses started to take advantage of a new opportunity that the internet offered: communication and interaction with and between the users. It enabled them to share files, discuss and share ideas, and finally to create together. Through such communication and collaboration an important, internet-based phenomenon known as “crowdsourcing” emerged. Outsourcing certain activities to a crowd allowed businesses to mitigate difficulties associated with two characteristics of knowledge: its wide distribution and so-called stickiness [von Hippel, 1994]. The lowering of costs associated with using the crowd gave rise to various crowdsourcing initiatives: crowds are now used to solve scientific problems, provide services, create content, and so on.

Crowdsourcing has gained increased attention as a new way of improving competitive advantage. On the one hand, it reflects real processes, especially globalization, changes in innovation environment, and IT-mediated technologies, which created new conditions to use of its resources [Roszkowska, 2017].

These technologies are physical and are also social ones that are enabling new ways of collaborating and managing activities across geographies, including emergency and management ones. On the other hand, crowdsourcing is perceived as an uncertain and risky process that inevitably accompanies the emergence of new concepts and their differentiation and is a subject of innovation in itself, for instance, business model innovation, open innovation business models, and one of its form, the crowdsourcing business model (CBM). The latter reflected a new approach to innovation. In these concepts, the process of exploitation of external knowledge that was acquired with internal one was combined. As firms posses and use different types of internal and external knowledge, there are different types of crowdsourcing and continuous process of differentiation of CBM among them. This pertains to broadcast search, crowd motivations, or crowd definitions [Amrollahi, 2015; Waldner and Poetz, 2015; Li, 2016; Massa et al., 2017] and types of platforms and generations. Each of them have different roles and characteristics [Estellés-Arolas and González-Ladrón-De-Guevara, 2012; Prpić et al., 2015; Prpić, 2016]. Also, we have other markets, including product ones, where heterogeneous firms function. The difference between them are product types and how modern they are, inner resources (according to the market they function in), ability to absorb external resources, their market placement, and how they compete, which encompasses how they add value. They also differ in terms of how they are structured, how long the market has existed, and their ability to be dynamic. As a result, there are different models of crowdsourcing and these vary in terms of their form, motivation, and the ability to use new forms of inner resources, including crowd knowledge and their result.

Crowd motivation and definition have received considerable attention from scholars, but still product and content crowdsourcing markets, an area where the crowd designs, develops, or selects products or content that are later sold, are still underestimated. These initiatives draw on crowd intelligence or crowd knowledge to create products or contents. A question arises: what similarities and differences do business models based on crowds utilized by different companies that function on different markets have? How do these models differ in terms of motivation, building blocks, including customer segments (CS) and relationship, value propositions, channels, key activities, resources, and revenues stream?

The aim of this paper is to provide an understanding of content and product markets’ crowdsourcing business models and show their differentiation. It could help different firms build similar models or compete against them or perhaps be a source of inspiration. We aim to systematically compare the similarities and differences of various types of CBM of some companies across a wide range of criteria to address the gap in the literature and bring about a more nuanced understanding of models. The business model definition and the framework used to analyze the business models are both based on Osterwalder and Pigneur’s [2010] work. We describe crowdsourcing business models of three product and content crowdsourcing companies and analyze and compare their business models’ key segments and points to some common features.

We start first, with literature review on business models, business model’s innovation, open business model innovation, and crowdsourcing business model. Next, description of the methodology that is used in analyzing the business models, the Business Model Canvas. We choose three cases of crowdsourcing business models of three various companies to ensure a degree of diversity. They are presenting diversified examples of the same business model category. A discussion of the results and conclusions are drawn in the end.

Starting from the mid-1990s, business model (BM) has become an important concept and field of research. In the literature, one can find reviews of literature on evolution of the use of the term. Zott et al. [2011] review of the literature published between 1975 and 2010, Hossain and Kauranen [2015] between 2006 and 2014, while Massa at al. [2017] between 2010 and 2015. They compared 71 definitions of business models and introduced their own taxonomy.

BM is seen as ‘the business logic of the firm’ [Teece, 2010; Hienerth et al., 2011; Velu and Stiles, 2013], ‘the platform which connects resources, processes and the supply of a service’ [Nielsen and Lund, 2014], ‘the way firms do business, creating and capturing value within a value network’ [Roome and Louche, 2016]. It shows how firm works¸ creates value and captures its part. A general definition of a BM: ‘stories that explain how enterprises work’ provided by Magretta [2002, p. 87] does not offer precise guidelines as to what a BM encompasses and how its analysis should be approached. Amit and Zott [2012, p. 42] tackle this matter from a different, more detailed perspective, stating that a business model is ‘a system of interconnected and interdependent activities that determines the way the company “does business” with its customers, partners and vendors. … a business model is a bundle of specific activities—an activity system—a conducted to satisfy the perceived needs of the market, along with the specification of which parties … conduct which activities”. BM is seen as a key component of an organization’s success.

The activity-based view of BMs captures its essence and suggests two sets of design aspects: design elements and design themes [Zott and Amit, 2010]. Design elements describe the activity system’s architecture of content, structure, and governance. Design themes coordinate and connect the three design elements and provide more detail on the activity system’s four crucial value creation drivers.

In the paper, we use the definition proposed by Osterwalder and Pigneur [2010, p. 14]: “A business model describes the rationale of how any organization creates, delivers and captures value.” Baden-Fuller et al. [2008, as cited by Casadesus-Masanell and Ricart, 2010, p. 197] provide a very similar definition of a BM: “the logic of the firm, the way it operates and how it creates value for its stakeholders.” These two definitions seem to leave most room for maneuver while grasping the core of the notion. Contrary to some other definitions, this one can be used in analyzing every business, not just a particular type.

Although business model is the organization’s logic for creating value, firms do need a change, improve model over time to remain profitable. Adjustment of an organization’s BM for creating value and innovate new models to beat competition seems crucial. Both practitioners and scholars agree that companies which choose to take advantage of these changes to produce new business models (business model innovation, BMI) prove to grow the fastest. In order to answer to changing external environment, firm must continuously change, reconstruct or innovate its business model. So, business model is dynamic phenomenon, could be innovated enabling new ways to create and obtain value and possess flexibility. Dynamic perspective on BM leads to establishment of BMI. It is “a boundary spanning process that must involve connections and joint innovation with external partners” [Ho et al., 2011, p. 663], and “a process that deliberately changes the core elements of a firm and its business logic” [Bucherer et al., 2012, p. 184]. BMI is creating whole new bundle of customer value. In BMI value chain architecture is revolutionized in new ways: from new product development to new delivery. It goes beyond process and product offering [Amit and Zott, 2012], or creating a new technology. It results in shifting of the focus to creation of the new system and broader network, i.e. beyond the entity of the firm: its customers, stakeholders and shareholders.

BMI is central for different types of firms as it also represents a new dimension of innovation. It can complement product or process innovation or it can occur independently from them. As it is a “distinctive type of innovation” [Bucherer et al., 2012, p.193] the concept of BMI broadens the boundaries of traditional meaning of innovation [Massa and Tucci, 2014]. However, it has some similar characteristics to other forms of innovation, for example, it can be incremental or radical as well as open or closed [Vanhaverbeke and Chesbrough, 2014, ch. 3].

In 2006, Chesborough launched a new perspective of analysis of BM called open business model. The focus of this concept is the role of the network. When it is combined with innovation, or when open innovation is included, this model takes a form of open innovation business model as opposed to closed innovation business model. As there are many forms of networks and ways of utilizing them, open innovation business model can take many forms. One of them is crowdsourcing business model [Marjanovic et al., 2012; Wikhamn and Wikhamn, 2013].

Crowdsourcing, a term coined by Jeff Howe [2006, p. 5], describes a BMI where a job traditionally performed by an agent is outsourced by making an open call to a large, undefined network of people. A company broadcasts a problem online, receives potential solutions posted by individuals, provides rewards for winning solutions, and mass produces the idea [Brabham, 2008a]. The information technology (IT) has enabled its expansion as companies can now ask IT users for help.

Equally important for the existence of crowdsourcing is the motivations’ driving individuals to participate. Sometimes financial rewards are an important motivation, as in the case of Innocentive, where the company provides financial prizes to researchers who provide solutions to posted problems. However, individuals often participate for other reasons, for instance because performing a certain task gives them a sense of enjoyment, provides them with an opportunity to socialize with others, or gives them a feeling of glory and recognition [Malone et al., 2010]. There are many studies that describe crowd motivations [Muhdi and Boutellier, 2011; Zheng et al., 2011; Dahlander and Piezunka, 2014] or in individual crowdsourcing companies [Brabham, 2010 and 2008a]. Empirical research underlines other area, for instance, factors influencing the originality of crowdsourcing ideas [Franke et al., 2013], process of idea generation [Schweitzer et al., 2012], and the outcome of crowdsourcing [Nishikawa et al., 2013].

The name crowdsourcing is formed from two words, “crowd” making reference to the people who participate in the initiatives, and the word “sourcing,” referring to a number of procurement practices aimed at finding, evaluating, and engaging suppliers of goods and services. This concept is linked with the wisdom of the crowds and collective intelligence. Surowiecki [2004] claims that under certain conditions groups are even smarter than the smartest individuals within them. He names four conditions that must be met for a crowd to be “smart.” The underlying idea of the wisdom of crowds is that if one asks a “smart” crowd to estimate or predict something and then averages the answers, the errors that each of the individuals makes will cancel themselves out. Crowdsourcing harnesses the wisdom of the crowd through IT to produce goods or services superior to those that could be produced by smaller groups. Waldner and Poetz [2015, p. 10] underlined that “crowds may possess relevant knowledge for BMI,” especially for delivering and capturing value. Companies use crowdsourcing to tap the wisdom of crowds and support the creation of collective intelligence. However, crowdsourcing can be sponsored and managed by the organization, which issues the task and assesses the results and whose purpose is to generate profit for the company [Kavaliova et al., 2016]. Hence, to create and value crowdsourcing, a company has to innovate its BM, that is, prepare crowdsourcing BM (CBM).

In the literature, there are more than 40 definitions of crowdsourcing [Estellés-Arolas and González-Ladrón-De-Guevara, 2012] and new definitions spring very often. The most popular is the notion that crowdsourcing is the IT-mediated engagement of crowds for the purpose of problem-solving, task completion, and idea generation and production [Howe, 2006; Brabham 2008a]. This tool has gained popularity [Hossain and Kauranen, 2015; Poblet et al., 2017]. Along with its growth crowdsourcing, another IT-mediated technology, for example, the sharing economy is rapidly being developed [Taeihagh, 2017].

Crowdsourcing is continuously evolving and a variety of its forms are emerging [Prpić 2016; Prpić and Shukla, 2016]. It can be carried out in two ways: by “proprietary crowds” that organizations foster through their own in-house platforms or by using “third-party crowdsourcing platforms” that provide the required IT infrastructure and “built-in crowds” as a paid service. Apart of external crowdsourcing through contests, internal crowdsourcing is used and is a part of an overall innovation capability for the company [Smith et al., 2017].

In the literature, crowdsourcing practices are divided into various types and forms. Different types of crowdsourcing represent different configurations of hardware, software, and people as IT-mediated crowds (such as virtual labor markets, tournament crowdsourcing, and open collaboration and as a new generation: crowdsensing, situated crowdsourcing, spatial, and wearable crowdsourcing]. One of the most popular taxonomies is done by Brabham [2011]. He categorizes these initiatives into four groups: knowledge discovery and management (the crowd is tasked with finding and collecting existing knowledge that is spread out over the internet), broadcast search (the problem that an organization defines and the crowd members find the solution), peer-vetted creative production (an organization asks the crowd to create and select creative ideas), and distributed human intelligence tasking (the crowd analyze vast amounts of information).

Another approach introduced Amrollahi [2015]. He selected 11 phases: conceptual design, participant selection deals, technical design (how crowdsourcing should be performed), the communication stage, invitation of the crowd, idea/task entry phase, the start of the interaction of the crowd with the crowdsourcing system, coordination of the crowd, idea/task revision phase (allows to remove any possible error from inputs), evaluation of entries, granting the award, and process evaluation and documentation of the lesson learned from the crowdsourcing project and future actions to implement the results of the crowdsourcing. However, this taxonomy does not include all crowdsourcing initiatives, for instance, crowd services such as GeniusRocket or uTest. Ghezzi et al. [2017, p. 9] reviewed 121 articles on crowdsourcing (published between 2006 and 2015) and investigated crowdsourcing from the input, process, and output framework. Providing formal structure for the findings available in the academic literature, they show the main antecedents, components, and outcomes of this process. In the literature, there are other forms of crowdsourcing based on different criteria: Busarovs [2011] introduced rewards criteria, while Li [2016] selected other forms. Täuscher [2016] names the challenges organizations face: corporate goals, the costs associated with crowd creation, the split of value added between crowd users, indirect and direct effects, and the proposal of solving the problems. However, Wilson et al. [2017] investigated failed crowdsourcing projects and pointed toward the types of risk that companies are likely to have.

Dawson and Bynghall [n.d.] suggested to distinguish between eight crowdsourcing business models and exemplified them by referring to their use by different companies:

In traditional company’s product design or new product generation is either done in-house or outsourced. If the design or creative phase is outsourced to a large enough crowd, some superior designs will prevail. Although the winning designs are usually awarded with cash, the design stage is still cheaper than keeping it in-house. What is more, the crowd is often tasked with choosing the winning design, thus making sure the winning designs actually sell, and indicating the potential level of demand for the good. This also lowers the need for market research. This approach is useful when matters of taste play a large role in choosing which products to introduce to market.

Chanal and Caron-Fasan [2010] analyze crowdsourcing experiment based on strategic analysis of a start-up Crowdspirit_. It outsources the entire R&D process to a community of designers and users in the domain of consumer electronics. They used the scenario-based approach to challenge the Crowdspirit business model and in a context of crowdsourcing it is necessary to distinguish between value creation (performed by the community) and value capture (should be made within the business model). They also suggest a new model of incentives for innovation activities called “collective private.” Smith et al. [2017] reporting on the efforts of EMC, a global leader in data storage system and clouding computing focused on intrafirm crowdsourcing of a large company.

In order to study and compare different business model based on crowds we use the Business Model Canvas framework developed Osterwalder and Pigneur [2010]. It shows the rationale of how a company intends to capture and create value. There are two main reasons why this framework will be applied in this paper. First, the selected nine areas of activities together provide a full view of an organization, since they cover the four essential areas of a business: customers, infrastructure, the offering, and the financial logic. Second, this framework has already proven successful in describing one crowdsourcing business [Innocentive] and can be used to compare businesses.

We concentrate on product and content crowdsourcing markets, where the crowd designs, develops or selects products or content, which are later sold. Content and product markets is a distinct category of crowdsourcing practices. It involves the outsourcing of design, creation or development of content or products to the crowd, and selling those mostly over the Internet. This category somehow differs from the rest as it draws on global, distributed creativity of users to create physical products or content. Moreover, it has real impact on the incumbent industries. For instance, iStockphoto is directly competing with professional photographers, in result lowering their wages [Howe, 2006]. Examples of crowdsourcing initiatives that belong to this category include Threadless, DesignbyHumans, I vote for art, Minted, iStockphoto, Shapeways, Quirky, and Ponoko [Dawson and Bynghall, n.d.].

Using the Business Model Canvas framework and based on criteria of diversity of firms’ characteristics and accessibility to information and data, we selected three firms (as case studies). The cases have been chosen to ensure a degree of diversity: two companies produce physical products, while one produces content; two generate considerable profits, while one has yet to cover its costs; one offers members the opportunity to earn their living, two do not as they differ in terms of motivations of crowds. One of them revolutionized the market on which it operated, forcing a strong drop in prices, when the other two prices did change but they changed the approach to the way of creating products. These companies differed in terms of the complexity of production processes and dynamics of growth. These case studies are mostly seen as presenting different examples of the same business model category.

Since both the term crowdsourcing and crowdsourcing companies are relatively new, this paper draws on mostly online sources, namely case studies, scholarly papers, magazine and press articles, analyzed companies’ webpages, interviews with companies’ founders (accessed mainly via YouTube), other videos, and e-mails.

The Business Model Canvas offers “a rigorous approach to structuring BMI efforts” and [Waldner and Poetz, 2015, p. 12] uses graphical framework, which covers nine blocks.

The nine building blocks of the Business Canvas are following:

Putting these nine building blocks together creates the Business Model Canvas, which allows for the analysis of a company’s business model.

Following the description of the business models, this paper identifies key segments in each company’s business model. The building blocks (following Osterwalder and Pigneur’s [2010] framework) are given in brackets using the abbreviations used in Section 5 of this paper.

The key building blocks of Threadless business model have been identified (Figure 1) as: customer relationship (CR), community catering (KA), brand (KR), community (KR), key partners (P), and channels ©. These can be presented using two mechanisms (Figure 2).

Figure 1

Figure 2

The customer relationship can be best described using three words: transparency [suggested by Lakhani and Panetta, 2007], fun, and excitement [proposed by Brabham, 2010]. Threadless, like many crowdsourcing companies, is transparent to avoid any backlashes from the community: even increasing the frequency of newsletters had to be accepted by the community before being introduced. Fun and excitement are another important aspect: many Threadless users enjoy participating because it brings them joy. The second building block, community catering (KA) takes on many forms: ensuring that designers have inspiration to design, keep on wanting to be part of the community, and that the community still wants to help the company grow, and so on. These two building blocks have two consequences: “love of community” and “brand community.”

The customers value the brand and “love” being a part of community because they find pleasure in interacting with others: working together on designs, discussing changes, waiting for the winners to be announced, and so on. They feel that Threadless offers them this opportunity—the company considers them an important partner by asking for opinions before making any changes to the website. These two effects, love of community and brand community, impact two other areas: brand (KR) and community (KR). As a result of the two effects, the Threadless brand has significant value; a competitor who would try to imitate the company would have to build their own brand first. Community is also the result of the two phenomena; the two effects make the community exist, be motivated to contribute, and “love” Threadless.

The second mechanism occurs with another building block: key partners. Both the retail partners and design partners increase the number of channels of interaction, nurture the community, increase sales, and serve as a marketing tool. All these consequences have an interesting outcome: they reinforce the network effect. The more channels there are, the more customers are reached, and the more designers want to join; the more the community is nurtured the more designers want to join, and so on. A larger number of talented designers produce better designs, which increases the number of people who want to participate.

iStockphoto’s business model logic is different. In this model, there are four key building blocks: customer relationship, community nurturing (KA), community (KR), and brand (KR). Community nurturing takes place from the beginning: if the photograph does not pass the screening process, the photographer receives feedback. By using feedback, tutorials located on iStockphoto’s webpage, and the forum (together these form the “support” relationship, part of CR), the photographers improve and start passing the selection process. As a result, iStockphoto creates its own pool or network of photographers, both amateurs and professionals, which makes the community its key resource (Figure 3). As the community learns and improves their skills, the iStockphoto brand is enhanced: the quality of submitted photographs improves and the company is probably seen as more professional. This is particularly important as photographs need to be checked for legal issues: if the brand is reliable, customers are not reluctant to buy the photographs.

Figure 3

There are five business model key building blocks in the case of Quirky: community nurturing (KA), channels (C), corporate culture (KR), customer relationship (CR), and community (KR) (Figure 4). The channels, for instance, Eval or brainstorming sessions, are meant to engage the community in the process of bringing a product to market. Quirky engages this community through these channels using its corporate culture, which emphasizes doing almost anything to bring the product to market. The channels and the corporate culture together nurture the community (KA) by making it engaged in the whole process. This, in turn, creates the relationship between Quirky and the community (CR), which is similar to that of Threadless: transparent, fun, and exciting. Community nurturing also has an impact on the community itself: it creates a community with certain traits that differs from other communities, for instance, that of iStockphoto.

Figure 4

Although key partners are not included in Figure 4, they are also a key building block: they allow products to be sold to the mass market, enabling Quirky to reach a broad customer base.

Comparing the three business models shows some commonalities between them (Table 7).

These business models have common building blocks: first, the key resource is always the community. However, the community is not a largely random crowd: it has been built over a period of time and holds certain characteristics. It is worth noticing that it does not always fulfill Surowiecki’s [2004] “smart crowd” criteria, for instance, iStockphoto’s community is quite homogenous. Second, another key building block is customer relationship: the result of such close interaction with the crowd is necessity of determining the type relationship that needs to be formed. However, the characteristics of the relationship differ between the cases: iStockphoto’s relationship can be described as transparency and support, while Threadless’s fun and transparency. Third, the key common activity is community nurturing with the main goal of motivating community members to participate.

Outside the Business Model Canvas other commonalities became evident:

This study illustrated the commonalities between content and product crowdsourcing business models. Although the business models had some differences in their key building blocks, three common segments were also captured. Four other themes, relating mostly to the elements of fun and the feeling of creating something, also surfaced.

The paper has implications for managers both employed inside crowdsourcing ventures and outside. Understanding crowdsourcing business models and identifying their key building blocks give those involved in crowdsourcing initiatives insight about factors crucial for their company and point to where they need to concentrate their resources and time. Crowdsourcing start-ups are also given a sense of direction regarding the various choices they can or need to make.

Understanding crowdsourcing business models has also implications for those who compete with crowdsourcing ventures, for instance, incumbent firms in relative industries. They can recognize the various strengths and weaknesses associated with these crowdsourcing business models and use them to their advantage. Also, when incumbents want to build their own communities, they can realize the cost associated with that and learn how to build such communities.

Companies wanting to use crowdsourcing methods need to remember that there are certain consequences of using crowds. First, having crowds as a key asset implies a necessity to take care of the crowd, to nurture it: letting it simply be neither builds a community nor sustains one. Second, such an interaction requires developing a specific customer relationship. Third, a barrier to entry, network effect, must be considered before launching a crowdsourcing initiative. Fourth, crowds need an element of fun or excitement to blur the line between fun and work. Fifth, crowdsourcing is often not just about the end product, but more about the “experience.” Mechanisms must exist to ensure that community members can benefit not just from introducing the end product, but also from interacting with each other. Sixth, the “joy of creation” might induce crowd participation, so tasks that have potential to wake such feeling should be considered to be given to the crowd.

To expand the base of knowledge on crowdsourcing initiatives, further research could consider applying a different framework, for instance, Amit and Zott’s [2012], which is tailored to e-businesses. Viewing the business models from a different perspective could provide additional insights and show new mechanisms. Comparing product and content business models to incumbent firms in their relative industries could also provide additional insight regarding where the strengths and weaknesses of such business models lie and propose how to use that knowledge. Another interesting area of research could focus on determining whether the “joy of creation” or the elements of fun and transparency are also driving other crowdsourcing ventures.

As with any other study, ours is not without limitations. The presented analysis is based on three case studies, and the methods, model that we put forth in the paper, are surely not exhaustive. Although the case studies were chosen to be diverse, the findings might be different when one analyses different case studies. Similarly, the cases were analyzed using a specific framework. However, each framework has its advantages and disadvantages. Although Business Model Canvas is useful to analyze and compare businesses, it has certain drawbacks for analyzing crowdsourcing ventures. For example, it doesn’t cover crowd motivations, which allow to understand the reasons why these companies function, and how they function. Similarly, it doesn’t easily capture any mechanisms, such as brand community, network effect, or switching costs. In this paper, this is evidenced by the reflections that emerged during analysis, but are not visible in the Business Model Canvas itself. We feel that our recommendation will only get better once more engagement research join our attempts at presenting various business models based on crowds.