When will 'crowdsourcing' NOT work?

This idea of seeking and collaborating with the community to seek solutions to issues and problems are back in vogue, what with the availability of web tools and ubiquitous digital connectivity that makes such an endeavour no longer the sole domain of the experts, or those presumed to be experts.

Wikipedia defines crowdsourcing as 'the act of sourcing tasks traditionally performed by specific individuals to an undefined large group of people or community (crowd) through an open call.'

Picture taken from: http://bit.ly/pGzjO0 

But with all the hoopla about the effectiveness of crowdsourcing, I do think that just like in most tools, there are reasonable limitations to leveraging on it as a hammer to nail in your problems. Perhaps these 5 are good key points of consideration for a start:

1. Size of crowd: Statistical textbooks dictates that there is a minimum amount of size that a sample should be taken from in order to validate any data obtained. Similarly, there ought to be a minimum sample size of sorts for a crowdsourcing endeavour to be deemed effective in its execution, notwithstanding its results, which I will point out a little later. And with the ease of digital connectivity, this might be one of the least hurdles that a crowdsourcer (versus the crowdsourcee) has to worry about, unless the specification calls for a certain degree of confidentiality, and the likes.
2. Levels of expertise: Perhaps one main motivation for a crowdsourcer to seek ideas from the public is the trust that he has with regards to the quality of the ideas and proposals that he would get. Although debatable, I do think that there ought to be some minimum level of cognitive familiarity with the issues or problems that are being crowdsourced. Though there is no necessity for all members participating in the crowdsourcing activity to be experts in that specific given field, there is perhaps a need for one to be at least intellectually aware of the issues at hand, and perhaps be able to give alternative ideas/proposals, bearing in mind the main objectives of the exercise. Within the community of crowdsourcees, there should also be that level of respect for alternative viewpoints, if well-justified and logical in its arguments. And in fact having multiple experts should be the order of the day, due to perhaps the ability for such a community to be able to enrich the crowdsourcing activity, much like how perhaps, the design firm IDEO, equips her design team for any design projects that she embarks on.
3. Motivation to contribute: At a personal level, although much have talked about crowdsourcing, but the motivation for one to participate in such an activity must also be something that a crowdsourcer be mindful of. Motivation, though not necessarily in the forms of something tangible, is a necessary part of the human natural psyche to act or do something, and likewise, there should be some carrot dangled for participation. A few models and sites are already available that leverages on some points/rewards system for contributors, but perhaps more could be done to attract participants to assist in contributing issues, especially of the social kind. Altruistic reasons aside, sometimes the economics of ideas and expert groups do not necessarily come in cheaply either.
4. Independence of contributors: Perhaps this is one of the primary hurdle that a crowdsourcer should look at in greater details. No doubt one couldn't really be placing in specifications to dictate the level of independence of the crowdsourcee in the whole scheme of things, but having respondents that do have a direct bearing on the outcome of the exercise is not necessarily a bad thing either. But of course certain guidelines and some degree of awareness should be put in place to create that level of 'independence of ideas' regardless of your dependency of the outcome of such an exercise.
5. Follow-Ups: Lastly, and perhaps more importantly, is the follow-up activities that such an exercise would entail after the dust has settled. Who would do the following-up? What is the timeline for such follow-up activities? What ideas or proposals would be followed-through, and which ones would be shelved or KIV'ed until further notice? Transparency, to a certain degree, is indeed the order of the day, especially to crowdsourcees, to give that level of affirmation that something is being done about their proposals. And this is especially important if the issues at hand are something that the respondents are directly being impacted upon, or if the crowdsourcer is looking at more of such activities in the future.