Lukas Biewald – CrowdFlower 2 of 2

“Be relentless about closing deal opportunities.” CrowdFlower’s on-demand crowdsourcing platform helps businesses complete large data-heavy projects fast.

[musical intro]

Matthew Wise: Hi this is Matthew Wise with FounderLY. We empower entrepreneurs to have a voice and share their story with the world, enabling others to learn about building products and starting companies.

So I am very excited today because I’m here today with Lukas Biewald, the founder and CEO of CrowdFlower. CrowdFlower is a leading labour on demand platform for easily getting work done.

With that said Lukas, we’d love for you to give our audience a brief bio.

Lukas Biewald: Sure, thanks Matt. My background is I’ve got an undergrad and a master’s degree at Stanford in computer science and math. From there I actually didn’t intend to be an entrepreneur. I worked at Yahoo Research for a few years and had a great time.

I came out of that and I worked at Powerset which was a search start-up a couple years ago. I had an awesome experience there. While I was there we had this really, really big need for crowd sourced work. We had to collect hundreds of thousands of judgements saying if a search result was a good search result or a bad search result. So we put it on a site called Amazon Mechanical Turk that a lot of your viewers have probably heard of where you can post jobs that are really, really short and anyone can come on and do them.

I had this awesome experience. I had this thing that I was expecting to take six months, I got done in, like, a day or two. So I went around to all my friends and I said, ‘Guys, you’ve got to use this service. It’s such a useful service, it’s this really powerful tool that helps businesses do things they were never able to before. It helps start-ups outsource things that would have been too hard for them to outsource in the past. And it helps people that might otherwise have trouble finding work actually find useful jobs.’ So I really, really loved this tool but I found that it was hard for a lot of people I talked to to use the tool. Because it was really designed for engineers, it was really low-level.

So what we made here at CrowdFlower is kind of an enterprise version of that. So you can take some important part of your business process and you can send it to the crowd through us, so we’ll put jobs on Amazon Mechanical Turk and 20 other platforms where people can come in and do simple tasks, and we’ll make sure that the quality is really high, we’ll make sure that the throughput is really consistent and good and we’ll make sure that the cost is really consistent and not fluctuating.

So that’s a model that’s worked really well for us.

Matthew: What makes CrowdFlower unique? Who is it for and why are you so passionate about it?

Lukas: Sure, I mean, so there’s kind of two people it’s for. So there’s the customers and part of why I’m passionate is the customers. Because what we allow customers to do is scale up and scale down the amount of work they need to get done. So say you want to, say you’re a big directory of businesses, there’s actually lots of these directories out there, there’s direct mailers which aren’t so exciting, there’s mapping services like MapQuest and Google Maps.

So say you have this big list of stuff and you want to check that all those locations are accurate, it’s actually really hard. The only way that we know today to check these locations is actually having people check online or call the locations and say, ‘Hey, Joe’s Bakery are you still there?’ and these places move all the time.

So what happens is, what’s happened in the past is it gets outsourced to India and you get thousands of people that sit in a room all day long, actually in the middle of the night there, so it’s daytime here and just calling these places, checking.

So our model, instead, says you can send us any number of these places that you want. You can send us one, you can send us a million and we’ll post these online, all over the web. And so anyone that wants to login can come in and do this simple task and we’ll pay them a small amount of money depending on the complexity of the task.

What that means to a business is they can effectively hire 10,000 or 100,000 people in five minutes. Not a month, right, like it would take with outsourcing. Not like five days. But literally like, right now. We could be talking in this interview and if you want to transcribe what I just said, you can send that out. In fact, you could get all your videos transcribed by sending it out to our service and it would get done almost instantly.

We actually do over three man years, person years of work every day. That’s 24 hours a day, that’s on weekends, holidays. So we’re allowing customers to scale up and scale down simple or repetitive jobs in a way they never could before. And we’re having them have their own employees do jobs that are more interesting. They’re not taking their employees and having them check every photo that got uploaded to make sure it’s not offensive, which, probably many of your viewers with their start-ups can related to. Like, looking through content, making sure it doesn’t violate terms of service. So I really love that we keep our customers super happy. So that’s kind of one part of the business I love.

The other part of the business that I love is that we’re enabling anyone to have access to a useful job. So if you go to you can’t do work on that site, but if you go to any of our many partner sites you can sign up and you can immediately be doing jobs and it’s super fair.

We can’t discriminate based on age, we can’t discriminate based on gender, we can’t discriminate based on race, we can’t discriminate on anything because we don’t ask you what your gender is. We don’t ask you what your college degree is. We’re judging the people doing the work completely based on the actual work that they do.

I feel really passionately about this, I think it’s going to lead to a more efficient, better world to live in.

Matthew: Can you speak about some of the technology and market trends that currently exist and where you see things developing in the future for your space?

Lukas: Sure. I think that there’s sort of an exciting space called crowd sourcing and it encompasses everything like Wikipedia which is a non-profit foundation, to things like InnoCentive or X PRIZE where you have these big contests to do stuff, to companies like us.

And it’s like, why are these all called crowd sourcing? I mean at first glance it’s not really a market, for sure. It’s maybe not an industry because the customers are all so different. I think of it more as almost like a society, a societal trend in the sense that technology has let us connect to people really fast and trust people faster than ever before.

And so it lets companies make these really quick relationships with people that might do useful work for them. If you think about it, it’s so crazy that we made everyone drive to work, sit in a box, we buy them a separate computer, we buy them a separate bathroom and we don’t often let people work in the middle of the night for lots of types of jobs, right.

And we have all this work here that we need to get done and there’s lots of people out there that want to do that work but there’s just so much friction in terms of finding someone, trusting that they’re actually going to do a good job, trusting that they’re not going to take company supplies. There’s all this friction.

And the internet lets you connect people really fast and keep a record of did they do a good job or a bad job? So I feel like you have things like Wikipedia where you have people that aren’t necessarily college professors writing these really smart, academic articles that are often super, super high quality, better than Encyclopedia Britannica or whatever.

You have these things like X PRIZE where it’s often really surprising groups that win these prizes. They’ll set up with something like, help with the with the oil spill, they come up with way to absorb oil or come up with a way to launch a rocket into space. It’s often some crazy guy in his garage because we’re just saying, ‘Hey, anyone can have access to this prize and this challenge and do something useful for these companies.’

Then what we do, the stuff that we do is like, ‘Okay, I need my website translated into Spanish.’ Well there’s a lot of people out there that are bilingual that don’t necessarily have access to jobs. It’s kind of hard to find them if you’re just . . . I mean, the friction of finding them is hard enough that you might just pay an agency or hire someone full-time or use an employee but when you have our kinds of quality control technology, we make it possible that you can just post on the web and trust that someone will do a good job.

Matthew: Can you tell us who your co-founder is? How did you meet and how did you know they’d be a good fit?

Lukas: Sure. So my co-founder is actually at my last job. So he’s Chris Van Pelt and he’s an awesome guy. You should interview him too at some point. We met because we’re actually both really passionate about . . . we were working together at Powerset on this using Mechanical Turk and seeing if we could get work to be done.

He was actually a little bit harder to convince to leave a funded start-up for an unfunded start-up. But he’s been great. We’ve worked together for almost four years now.

Matthew: From idea to product launch, how long did it take and when did you guys actually launch?

Lukas: Well that’s a good question. I mean we’re sort of . . . I mean, we launched pretty fast. But it’s been . . . there’s been so many years of iteration. I wouldn’t say we launched and the customers showed up [laughs]. Our whole . . . we bootstrapped for almost a year, not because we wanted to but because we weren’t making something that people actually wanted [laughs]. So no one would fund us.

I think that, I sort of feel like the product has only been on the market in a way people would actually . . . in the way that it is today for about a year and a half. So it probably took us, like, two to three years to get the product to a point where people would actually pay real money for it.


FounderLY Photos

			FounderLY posted a photo:	Future of Education - FounderLY			FounderLY posted a photo:	Future of Education - FounderLY			FounderLY posted a photo:	Future of Education - FounderLY			FounderLY posted a photo:	Future of Education - FounderLY			FounderLY posted a photo:	Future of Education - FounderLY			FounderLY posted a photo:	Future of Education - FounderLY

Press Mentions #1

Press Mentions #2