Have we all been imbibing the cool aid? Are the likes of Wikipedia really crowd-powered?

In a recent well-argued article in Forbes – The Myth of Crowdsourcing – Dan Woods claims crowds don’t innovate, individuals do.

Crowds and uniquely talented individuals

“There is no crowd in crowdsourcing. There are only virtuosos, usually uniquely talented, highly trained people who have worked for decades in a field. Frequently, these innovators have been funded through failure after failure. From their fervent brains spring new ideas. The crowd has nothing to do with it. The crowd solves nothing, creates nothing.”

He goes on to point that what are often called crowdsourcing platforms really are virtuoso search platforms.

Apparently Dan Woods accosted Wikipedia-founder Jimmy Wales at a conference last year and asked him about how articles were created.

“He said that the vast majority are the product of a motivated individual. After articles are created, they are curated–corrected, improved and extended–by many different people.”

I agree with Dan Woods to an extent. Just like much of the sharing on social platforms is actually just egotistical self publishing, crowds are often driven by a few talented individuals. I have discovered brilliant individual photographers on Flickr, but you do have to wade through quite a bit of mediocrity first.

The LA Times’s experiment with a Wikitorial – an attempt to have a user-created and contributed editorial on the Iraq War – is proof of how the crowd can get it wrong.

“On Friday, the paper introduced an online feature it called a wikitorial, asking Web site readers to improve a 1,000-word editorial, “War and Consequences”, on the Iraq war.

Readers were invited to insert information, make changes or come to different conclusions.”

It did not last.

“A Los Angeles Times experiment in opinion journalism lasted just two days before the paper was forced to shut it down Sunday morning after some readers repeatedly posted obscene photos.”

Want to see something not very cool that sounds awful? Then look at MTV’s Amplichoir below. It’s part of a marketing campaign and billed as the world’s biggest crowdsourced choir. Users are incentivised to take part via a competition prize.

It screams fake, sounds horrid and its pastel coloured iPod-esque backgrounds look contrived. Mr. Woods I’m sure would agree that this proves his point. It does not work because there is no talented individual(s) to make something of it.

But my agreement with Mr. Woods only goes this far.

YouTube is full of bad user-submitted videos – and the odd good one , but as a whole it is collective effort. Most quality Wikipedia articles may be driven by an individual user, but the whole is a “crowdsourced” phenomenon.

And both YouTube and Wikipedia have been increasing mechanisms that make collaboration and reaction to others’ contributions possible. This allows us to feed off, incorporate and build on ideas.

Curveball! Kutimans splicing together of YouTube videos into fantastic new ones, is that not evidence of a crowd of virtuoso’s being used and orchestrated by a virtuoso?

Where the crowd’s contributions stop and the virtuoso’s starts is not always so clear cut.

Creation vs Evaluation

There are of course two kinds of ways to tap into collective intelligence. And perhaps that’s where Mr. Woods confusion arises.

The one – like Wikipedia and like Flickr is where people – yes individuals – create.

But there is another form. i.e. to evaluate existing ideas and creations – and this often happens anonymously.

It’s when we look at the power of collective evaluation – like with voting mechanisms, market prediction systems or systems like Google’s Pagerank (effectively a voting mechanism that counts links to predict web page importance), that we can see a more pure form of collective intelligence in action. Google does an amazing job of finding good websites based on our links.

In other words, where we use collective methods for large scale evaluation and not ‘just’ for ideation or creation we have more pure examples of ‘crowd’ intelligence. But even these lines are blurring.

Digg and the Starbucks and Dell idea platforms allow users to submit ideas, and others to vote on them. Eat your heart out Mr. Woods.

First, can I just say that I really don’t like the term crowdsourcing.

Why? Crowd to me sounds like just more jargon – a bit like oft used tribes. And Source? Well, this a sibling of that other contentious word – outsourcing. Many of the most successful platforms in this area – like Wikipedia – are not commercial in nature at all. “Crowdsourcing” is so much more than just a management strategy to cut costs.

We know “crowdsourcing” is powerful (our logo came via a “crowd”), but how do you fully harness this power? How do you decide what functionality to have to maximise contributions?

We have just been asked to help conceptualise and build a “crowdsourcing” platform. A platform for collective ideas generation. It should produce better creative ‘concepts’ and get the right people to execute them. Importantly – unlike similar solutions out there – it is different in that it would not be open to everybody.

So I’ve been looking at many online examples to see what’s out there in the ‘wild’. That’s important because as an MIT’s Sloan management review study – Decisions 2.0: The Power of Collective Intelligence – acknowledges: practise is still some way ahead of the theory.

Of all these platforms, the ones that have impressed me most for the purpose of what we are building are – in no particular order:

CrowdSpring – Focused on design (graphic, web, product), it already has a big community of designers and people requesting designs. The requester or buyer can interact directly with creatives and picks the winner. Anybody can be a buyer or a creative, and portfolios are public. A whole set of metrics are visibly published so buyers and designers can make decisions based on reputation. This reputation dashboard also serves to regulate behaviour.

Eyeka, a French platform, is a social network for creatives. It also allows brands to run design competitions. Each competition gets its own URL so the creative process and the selection of a winner becomes a marketing exercise in itself. The community picks the winners through a voting system. Anybody can join Eyeka.

InnoCentive is one of the most talked about collective platforms. Here seekers post sophisticated challenges they want solved – like a computational problem. The challenges are public and so are the profiles of the solvers. All challenges have monetary incentives. Anybody can join InnoCentive.

And then there’s the Dell and Starbucks Idea platforms built on top of Saleforce.com platforms. These two have been discussed in many good social media books, from Forrester’s Groundswell to Jeff Jarvis‘ What would Google do, and many more. For a reason. They are very active and work. Anybody can post an idea and others can comment or vote on it.

There are also some much smaller networks. London is the home of Radar Music Videos, a small social network for video directors, which also allows commercial briefs to be posted. You have to pay a small membership fee to access briefs and post videos. The network seems quite active, but alas it does not have that many briefs.

It is worth noting that there are many platforms out there that look like they are on their last legs, virtual tumble weed is blowing across interfaces barren of users.

I don’t need any convincing about the power of digital media to harness sharing of all kinds of things. So what did I learn?

Incentive is the key issue. People’s motivations as to why they take part in these things will determine the platform’s functionality, its mechanisms. As the above mentioned MIT study says:

“An application that taps into collective intelligence for improved decision making may be simple in concept, but it can be extremely difficult to implement. As with many systems, the devil is definitely in the details.”

The devilish details

The so-called devil’s in the detail. So what are some of these details?

How much control do you exert?

Jeff Jarvis says give up control, the users will take control and run with it – more often than not in a good way.

But do you give your ‘crowd’ the ability to choose winners (for example)? What if you don’t agree? If you manage to build a platform where a number of people have contributed to a creative approach, do you split the rewards between them? How? Do you let them decide? Should brands be able to interact with the community directly? These are just some questions we will have to face shortly.

Diversity vs expertise

You have to get the balance right. If your collective is not open, how do you choose participants? If it’s a small group of experts, how good will they be at evaluation (studies show diverse and large groups are better at evaluation – and I will blog about that tomorrow)? If your members don’t like each other, are they likely to stay?

Engagement

What motivates people varies wildly – the MIT report explains:

“Incentives such as cash rewards, prizes and other promotions can be effective in stimulating individuals to participate in activities like prediction markets, for which explicit rewards seem to matter greatly. With other applications — for example, submitting T-shirt designs to the Threadless Web site — cash rewards seem to matter less than recognition. Value-driven incentives can also be important. As the open-source movement, Wikipedia and other similar efforts have shown, participation in a community, the desire to transfer knowledge or share experiences, and a sense of civic duty can be powerful motivators.”

Another problem is keeping people engaged over time. And if your users are primarily incentive-driven and there’s too many competing for a limited pool of cash, then what?

So how do you measure success?

Well it obviously depends on the goal of your platform. But there is one important way to measure success. Is it being used?

“Engagement should not to be taken lightly. Indeed, for a large fraction of Decisions 2.0 projects that have flopped, the primary cause of failure appears to be a lack of engagement. Participants expect to be treated in a certain way and, more often than not, they also want the organizers of the application to be engaged as well.”

With a little uncommon sense, a little theory and many examples of what has worked and what not, you should have as solid a start as you can hope for.