Crowdsourcing - Typologies

From P2P Foundation
Jump to navigation Jump to search

see: Crowdsourcing, for general treatment


Introduction: Classification of Crowdsourcing Approaches

Katarina Stanoevska-Slabeva:

"In literature two basic approaches to classify crowdsourcing can be identified:

1) based on the type of task that is crowdsourced. A representative classification in this context is the classification provided by Howe (2008), and

2) based on the initiator of crowdsourcing. A representative classification according to this criterion is given by (Gassmann et. al. 2009). Both classifications are summarized below."

See: Classification of Crowdsourcing Approaches

  1. Classification based on task
    1. Crowdsourcing Idea Game
    2. Crowdsourced Problem Solving
    3. Prediction Markets
  2. Crowdsourcing typology by initiator
    1. Crowdsourcing initiated and supported by intermediary platforms.
    2. User initiated crowdsourcing
    3. Company initiated platforms
    4. Idea market places
    5. Public crowdsourcing initiatives


The fourfold typology of Emma Johnson

Emma Johnson:

"In his book, Howe lays out four types of crowdsourcing: collective intelligence, crowdcreation, voting and crowdfunding.

Collective Intelligence

The first type assumes that the masses are smarter than individuals. Spiceworks, an Austin, Texas-based software firm, built a free application that helps IT managers of small businesses manage all their software and hardware. Integral to the product is a feature that invites users to comment on its various components and vote on others' suggestions. The two-and-a-half year-old Spiceworks now has 500,000 users and the software is now in its 10th version--with revenue generated by advertising.

Make sure you listen to the feedback you receive. Spiceworks CEO Scott Abel said when they started out, it was tough to take seriously comments that contradicted years of experience and heaps of market research. "They're telling us the way things are in the real working world, and we were reverting to our old market research ways," he says. "Today customers expect to be listened to--and if they're not, they're going elsewhere."


The second type, crowdcreation, has been successfully used by companies including Threadless and 99 Designs, a site on which those in need of logo, business card or website design can post an assignment and fee, and designers submit designs for consideration. The contest sponsor then chooses one design and awards the fee. Mackenzie of Gradigio used 99 Designs and other crowdcreation sites.

"The big advantage was that we got a broad range of designs right up front," Mackenzie says. "Traditionally, you work with one designer, and you never know what you're going to get."

Make sure that you pay appropriately. Many of those who participate in this form of crowdsourcing are professional scientists, designers and other professionals. Free publicity and an ego boost are not adequate enough compensation to attract top minds.

Be prepared for an avalanche of information by creating an in-house mechanism to filter and sort information.


The third type, voting, collects public sentiment by asking individuals to give feedback on an existing idea or product--in the case of Spiceworks, software users vote on peers' improvement ideas, while TripAdvisor aggregates travelers' votes on hotels and cruises.

Use it as free market research. Take advantage of the comments and ratings by making suggested changes to your business.

Respond. In the case of review sites, log in to the sites as a representative of your business and send private messages to those making mistakes. Be careful about posting public replies, as that can have a shouting down effect that makes others afraid to comment if they fear they might receive a snarky reply from the business.

Encourage customers to post comments and reviews through messages on marketing material.


The fourth type, crowdfunding, refers to the public's willingness to finance projects they believe in: lending money to a microfinancing project, funding an independent film, sponsoring a piece of journalism, or the takeover of a soccer team, in the case of MyFootballClub's collective purchase of England's Ebbsfleet United Football Club for 700,000 in 2007.

Be transparent. People want to know where their money is going.

Show something for it. Whether it's a profit from an investment, or a meaningful story about how donations helped someone in need, show users results." (

Ross Dawson's six-fold typology

"Here are the six types of crowdsourcing mentioned in the article:

1. Distributed innovation platforms: "They find more than half the people that solve the challenges on Innocentive and these other distributed innovation platforms already know the answer. So why should they solve that problem again when they can find someone else who already knows the answer?"

2. Idea platforms: "These sometimes go under the guise of idea management software, but these are ones where people inside organisations – often – submit ideas or proposals for cost savings, or new products, or new services, or process efficiencies, and then they collectively assess and rate and vote on and select and evolve and refine and build on those ideas to become the innovation that will drive that organisation forward."

3. Innovation prizes: "Anybody anywhere can enter their own projects and ideas, others can vote on them and build on them and use the wisdom of the crowd to make them more effective, and from all of those submissions somebody wins a quarter of a million dollar prize."

4. Content markets: Threadless and Red Bubble are mentioned.

5. Prediction markets: "For enterprise software companies it is notoriously difficult to forecast sales. For many reasons, the sales pipeline that is put into CRM systems is often inaccurate. However, if you then ask the salespeople to predict what the sales are going to be for that quarter and you aggregate all of their opinions, you can get a far more accurate view of what the actual sales are going to be."

6. Competition platforms: DesignCrowd, CrowdSpring and Guerra Creativa are mentioned." (

In 'Getting Results from Crowds', Ross Dawson lists 20 variations:

  1. Service marketplaces Matching buyers and sellers of services.
  2. Competition markets Competitions awarding prizes to selected entries.
  3. Crowdfunding Donating to creative ventures, sometimes as a pre-sale.
  4. Equity crowdfunding Equity funding from many small investors
  5. Microtasks Markets for very small well-defined tasks
  6. Innovation prizes Prizes for single, defined innovation outcomes.
  7. Innovation markets Matching clients and researchers for innovation.
  8. Crowd platforms Software used to support crowdsourcing processes.
  9. Idea management Processes to propose, rank, and improve on ideas.
  10. Prediction markets Coalescing diverse views into collective forecasts.
  11. Knowledge sharing Sharing knowledge, experience, and insights.
  12. Data Gathering or refining data in specific domains.
  13. Content Creating media content.
  14. Content markets Enabling creators to sell their content.
  15. Crowd process Aggregation and added value to marketplaces.
  16. Labor pools Access to groups of specialists.
  17. Managed crowds Aggregated services provided by selected specialists.
  18. Crowd ventures Businesses conceived and managed by crowds.
  19. Citizen engagement Contribution to civic or government initiatives.
  20. Contribution Philanthropic fundraising and ventures.
  21. Science Contribution to scientific endeavors.

Patrick Meier's Allsourcing

Patrick Philippe Meier:

"the term “crowd” can mean a large group of people (unbounded crowdsourcing) or perhaps a specific group (bounded crowdsourcing). Unbounded crowdsourcing implies that the identity of individuals reporting the information is unknown whereas bounded crowdsourcing would describe a known group of individuals supplying information.

The term “allsourcing” represents a combination of bounded and unbounded crowdsourcing coupled with new “sourcing” technologies. An allsourcing approach would combined information supplied by known/official sources and unknown/unofficial sources using the Web, e-mail, SMS, Twitter, Flickr, YouTube etc. I think the future of crowdsourcing is allsourcing because allsourcing combines the strengths of both bounded and unbounded approaches while reducing the constraints inherent to each individual approach.

Let me explain. One main important advantage of unbounded crowdsourcing is the ability to collect information from unofficial sources. I consider this an advantage over bounded crowdsourcing since more information can be collected this way. The challenge of course is how to verify the validity of said information. Verifying information is by no means a new process, but unbounded crowdsourcing has the potential to generate a lot more information than bounded crowdsourcing since the former does not censor unofficial content. This presents a challenge.

At the same time, bounded crowdsourcing has the advantage of yielding reliable information since the reports are produced by known/official sources. However, bounded crowdsourcing is constrained to a relatively small number of individuals doing the reporting. Obviously, these individuals cannot be everywhere at the same time. But if we combined bounded and unbounded crowdsourcing, we would see an increase in (1) overall reporting, and (2) in the ability to validate reports from unknown sources.

The increased ability to validate information is due to the fact that official and unofficial sources can be triangulated when using an allsourcing approach. Given that official sources are considered trusted sources, any reports from unofficial sources that match official reports can be considered more reliable along with their associated sources. And so the combined allsourcing approach in effect enables the identification of new reliable sources even if the identify of these sources remains unknown.

Ushahidi is good example of an allsourcing platform." (

Nicholas Carr's fourfold typology of Crowds

Nicholas Carr [1]:

""Social production crowd": consists of a large group of individuals who lend their distinct talents to the creation of some product like Wikipedia or Linux.

"Averaging crowd": acts essentially as a survey group, providing an average judgment about some complex matter that, in some cases, is more accurate than the judgment of any one individual (the crowd behind prediction markets like the Iowa Electronic Markets, not to mention the stock market and other financial exchanges).

"Data mine crowd": a large group that, through its actions but usually without the explicit knowledge of its members, produces a set of behavioral data that can be collected and analyzed in order to gain insight into behavioral or market patterns (the crowd that, for instance, feeds Google's search algorithm and Amazon's recommendation system).

"Networking crowd": a group that trades information through a shared communication system such as the phone network or Facebook or Twitter.


Clay Shirky, who is also participating in the discussion, suggested a fifth crowd type for this list:

"Transactional crowd": a group used to instigate and coordinate what are mainly or solely point-to-point transactions, such as the type of crowd gathered by, eBay, Innocentive, LinkedIn and similar services. (I would think that contests like the Netflix Prize also fall into this category.)"

An extra suggested category:

"Event crowd": A group organized through online communication for a particular event, which can take place either online or in the real world and may have a political, social, aesthetic, or other purpose." (

Resource Crowd, "which is represented best by crowdfunding platforms such as Kickstarter, Kiva, and IndieGoGo. These platforms aggregate small amounts of money to accomplish a fundraising goal to complete a project. They allow for participation in the funding of projects that might be far out of reach for the average contributor while also providing the funding necessary for a producer that may not have been able to raise funds before such a platform. The power is in the aggregation of small contributions by a large crowd." (

Belsky and Kalmikoff's Crowdsourcing Business Models

"According to Belsky and Kalmikoff, the crowdsourcing definition needs to evolve, especially beyond the common misconception that crowdsourcing means access only to free labor. They mention three business models:

1) Crowdsource wisdom (or knowledge/expertise/skill), as with Wikipedia. 2) Crowdsource labor, as with Amazon’s Mechanical Turk, or traditional spec contests. 3) Crowdsource both wisdom and labor, as with Digg or Threadless. Keep the community active in the business." (

Task-based, employee-based, information exchange

By Derek Smith, Mohammad Mehdi et al.:

Task-based public crowds

Public task-based crowds perform a specific task or a set of tasks. There are few or no relationships between the crowd participants, who each contribute using their individual abilities. We identified seven articles in this literature stream (Table 1).

The field settings examined in this research include six intermediary companies offering a crowdsourcing service to customers or corporations and two companies that use internal crowdsourcing capabilities as part of their business model. The types of design tasks in this stream relate to electronics, product design, digital media products, T-shirts, graphics, advertisement, and websites.

Motivational drivers examined in these articles include: immediate financial payment (of varying amounts), skills improvement, enjoyment and fun (of varying type), and community-related motivations. Technology entrepreneurs involved in similar tasks could consider motivating their own crowds in similar ways. However, it is important to note that financial payment might not be the best way to motivate a crowd (Antikainen et al., 2010; Bogers and West, 2012) because other forms of motivation can be more important.

Employee-based crowds

Participants in a corporate, employee-based crowd are employed by the host company. We identified three articles in this stream (Table 2). Two articles examined large-company field settings: one was based in Switzerland (Muhdi and Boutellier, 2011) and the other examined a multinational corporation (Stewart et al., 2009). The third article (Hossain, 2012) is a literature review on motivation and incentives. The crowdsourcing tasks examined in this literature stream included internal idea generation and language translations.

Motivational drivers in these articles include immediate payment of rewards, such as peer recognition, career advancement, and professional development.

Information-exchange public crowds

This type of crowd includes participants seeking technical information as well as participants providing technical information, and these roles are interchangeable. Some tasks may also require creativity in addition to technical information.

We identified three articles in this stream (Table 3). Two articles examined field settings; one was an ideas-based community organized around a company-sponsored contest, and the other was a knowledge-based community anchored around online Usenet groups about computer programming and databases. The third article was a controlled experiment by marketing researchers who could manipulate the points system used to reward participants for contributions and thus shape crowd behaviour. In this last article, one point system resulted in more new ideas, while a second point system resulted in more ideas that built upon existing ideas.

Motivational drivers examined in these articles include access to technical experts to solve problems, learning, fun, and being part of a community." (