Publicyte Interviews Lukas Biewald CEO Of Crowdflower, Which Offers Crowdsourcing As A Corporate Service

Lukas Biewald may not currently be a household name in government circles, but he's getting there. The 29-year-old CEO of San Francisco-based Crowdflower has been making the rounds as a featured speaker in public and civic sector conferences like Tim O'Reilly's Gov 2.0 Expo and Andrew Rasiej and Micah Sifry's Personal Democracy Forum.

Publicyte editor-in-chief Mark Drapeau caught up with Lukas over lunch at Otto in Greenwich Village (NYC) for a chat about how his company's work use of crowdsourcing matches up with challenges facing public and civic sector organizations.

What is Crowdflower and how did it come about?

My co-founder and I were working extensively with Amazon "Mechanical Turk", an early platform for crowdsourcing simple tasks for a company called Powerset (which was later sold to Microsoft). Mechanical Turk gave us the initial idea that we could focus an entire company on delivering crowdsourcing services to clients. Fast forward a couple years, and we founded Crowdflower in 2007. At a time when most people barely knew what crowdsourcing meant, we started working with big data for enterprise clients.

CrowdFlower breaks down large repetitive jobs into small pieces and has lots of people work on them all over the Internet.  This lets anyone, anywhere get compensated for useful work. From some of my earlier work in artificial intelligence, I realized that many of the promises of artificial intelligence could be realized by using human intelligence on a large scale.

Where is Crowdflower based?

We're still in tech startup mode in San Fransisco, currently based in the eclectic and fun Mission District.

What has your experience been interacting with tech communities focused on the public and civic sectors, like when you've attended the Gov 2.0 Expo I co-chaired in 2010?

We've seen quite a growing interest in these different sectors, and we're actively looking for opportunities to explore different forums and to get Crowdflower into new verticals. For example, I'm going to a geo-spatial information conference soon in the Middle East. In another example, we've been talking with someone about crowdsourcing urban planning in Vancouver. And we definitely keep an eye on the big picture - I think initiatives like crowdsourcing the new Icelandic constitution are awesome.

I would add that the tech community focused on the public and civic sectors is really smart and motivated,but it's really small compared to the broader tech community that mostly mistrusts government.

Crowdflower's clients/customers tend to be large corporations. Would Crowdflower consider taking clients in the government or other nonprofit organizations?

In fact,we do have clients in the government and nonprofit sectors. And there's no question that Crowdflower is interested in doing more work in government, healthcare, and other public and civic verticals. And we're also involved in using our crowdsourcing platform for philanthropic causes; Check out Mission 4636, for example, which has done a lot to help post-disaster Haiti.

Large public or civic organizations often have a large group of fans or constituents that corporations don't have. This makes crowdsourcing even more relevant and useful for them. One of the challenges a corporation faces is motivating people; many civic sector organizations don't have that problem and they can even increase engagement while getting useful work done.

What's the biggest misconception about using crowdsourcing to solve organizational problems?

There's a lot of sketpicism about whether or not it actually works. We constantly have to "draw back the curtain" to show quality that the client is looking for. Clients really want to know, "if we do this, how do I justify this innovative approach?" We have to guarantee the scalability and accuracy that our clients demand from us. We need to get to 99% accuracy for our clients, and the difference between paying for 99% accurate versus 97% accurate is what differentiates us from our competitors. In order to do that, we have a larger labor source and more operational experience.

Will crowdsourcing be relevant during the 2012 elections? How?

Aboslutely. Crowdsourcing is likely to be seen in everything from sentiment analysis to doing background research on candidates and issues. Whenever you need large data sets in real time is where crowdsourcing comes in. This is probably going to be true even during the actual elections themselves.

What can people expect to see from you in the near future?

I deeply care about philanthropic and public-good uses of crowdsourcing, so I think you'll see more from us in the humanitarian sector - crisis relief work, etc.

Lukas Biewald is the Chairman of Crowdflower.