TED Conversations

This conversation is closed.

What makes a crowd-sourced (collaboration) project successful?

We see lots of successful collaboration projects, while we don't notice the ones that fail. Many projects succeed and many fail. What are these characteristics that make collaboration successful?


Closing Statement from Juan Calderon

I now have a better understanding of what a collaboration project should have. Successful Crowdsourcing (collaboration) projects have a goal with which the crowd has to identify. The project has to be organized in modules and have very clear instructions for the crowd to follow. The crowd has to be rewarded within the following categories (monetary, experience, and social/psychological. It is within the human nature to collaborate. Yochai's paper referenced within this discussion is amazing. The outcome of the conversation as was given to my class can be seen on this prezi http://prezi.com/eyb7vhlvidsu/crowdsourcing/

Showing single comment thread. View the full conversation.

  • thumb
    May 5 2012: Kind of black and white don't you think? Like normal society all the things being done on the internet are a myriad of different practices with different goals and background. I don't personally think there a specific approach towards the internet that differs from real life. The only thing that might change is the technology in use.

    So in my opinion a crowd-sourcing project would rely on good marketing and a vision that is attractive.

    There are so many different ideas about this, so why not make a debate. Then we can also start debate about the meaning of success.
    • thumb
      May 5 2012: Marketing and vision are absolutely important.
      Within your marketing you have to sell your project, but the project is not just marketing.
      There have to be some incentives for people to join, that not necesarily are related to you being aligned with the vision.
      I believe that there is some structure, some arquitecture that helps projects develop. I'm wondering about the characteristics that help projects be successful.
      • thumb
        May 5 2012: Then let me tell you about my experiences, as I do have some.

        With crowd-sourced I think you mean free labor right? The essential thing is a goal, without a goal there is no development. However picking out what the main goal is is crucial. It should be inspiring and promote a better future, the next thing you have to look for is short term goals. The next thing that is on your mind is: how big is our group? How many people are working on it actively without being paid until the end? A small group means freedom and close interaction, however the bigger the group becomes you will need to divide it up between departments. However, not separate them, direct interaction is still needed. So as organizer, when you are talking about a thousand people, you have to make sure it self-organized. Create a flow system so people know what things they have to complete, just like when you are learning a new trade. Instead of focusing on organizing the whole thing you now have the time to spend your time on the so called "positive deviants": people that will take action themselves because they seek a solution and do this without being recognized for their actions (say irrational). By putting them in the spotlight you boost the morale between the short term goals.

        With self-organizing I also mean the monitoring of tasks, but this can be done through software (if the resources are there).

        The next important thing is to not overflow the project with new people. People that are unaware of the progress being made. It is better to let outsiders have access to the development and any interested can still apply for the job. Though adding modules to the already existing project will bring in new people involuntarily. This is however not something bad. It requires the people working on the projects to see it through novice's eyes, which in my experience has always lead to solving some of the most crucial problems.

        I hope this is what you're looking for.
        • thumb
          May 6 2012: Thanks. This is what I'm looking for. It only rises more questions.
          There is the marketing, there is the vision or goal, and there is the free labour for which you have to establish the right conditions so that it "self-organizes". This is made by giving short term goals which have to be aligned towards the goal for the system to flow. And they should be added in a modular fashion.
          How do you keep the free-labour inside? There is some recognition and reputation involved. So you need some gamification? i.e. rankings, prices. money.
          How do you control? You have to monitor and you have to give some feedback so that the system doesn't go towards sommewhere else. Is this done by making each task small and simple enough?
      • thumb
        May 6 2012: I understand it rises only more questions, but like I said in the end there is a myriad of variables at play.

        However, when you organizational design it all relies on the product you want to bright forth and how many parties have a particular role in the whole scheme. Creating an IT company is fairly easy in contrast to a drug company. The bigger the thing gets the more levels you have to create. Now I am not talking about mid-field managers, these people play no beneficial role in my eyes. Not because of their position, but because of their credentials.

        Free-labour? Labour is almost never free, maybe it is voluntarily, but it is rarely free. Labour can only be free in a sense if the participating people already earn more than they need and do the work for free. So in the end it only like a serious hobby, rather than a job. And when it becomes a hobby you need to adjust your organization structure. More freedom, but in what way I can't say as it against depends on multiple factors.

        Gamification, yes it works, but only for short-term (activist) groups. I have yet to see it integrated in a matured long-term group. Though not impossible, I have never succeeded in such and I actually come from the gaming industry formerly. Which should be the least hostile to such measures. ::: On another note, gamification does lower the threshold for community interaction if applied on a large scale, and with it being on a large scale it can be made self-regulatory. A simple example of self-regulation is thumbing up/down. Not everyone will do it, but enough will to add value to the system.

        May I ask what this is for? If it is for a specific task I might be able to be more thorough.
        • thumb
          May 6 2012: Thanks.
          My friend asked me to help her by teaching a class next Tuesday. I'm working on the presentation which is full of examples. I started this conversation in order to help me organize my ideas.
          The presentation has many examples and I was missing the key points and a call to action. I want them to leave class with some simple rules of thumb or a nice summary of what makes it work.
          I appreciate your contributions, they have helped me organize the ideas.
          Any other comments or suggestions will be useful.

Showing single comment thread. View the full conversation.