Acquia, uTest, GeniusRocket, and Local Motors Gather to Discuss Crowdsourcing

Below is the transcription of the Ultra Light Startup's panel discussion on crowdsourcing business models that took place on Boston on May 4th. The panelists included:

Jay Batson - Co-founder of Acquia, a start-up commercializing open source Drupal .

Peter LaMotte - Marketing Director of GeniusRocket , a start-up crowdsourcing marketing and advertising content.

Doron Reuveni - Co-founder of uTest , a start-up crowdsourcing software testing.

John B. Rogers - Co-founder of Local Motors , a start-up crowdsourcing automotive design.

Moderator:

How can start-ups and companies of all sizes bring crowdsourcing into their business?

John B. Rogers:

So, there's two questions that I ask them first: Who's your customer? And, who's your crowd? If you can identify that as a business, and have some notion of what that is, then you can start to build a strategy around that. Because although many people ask that, they either don't have a crowd, or they don't have a customer - and its difficult to build a crowdsourcing model without those two things.

Doron Reuveni:

I think one thing that everyone needs to understand, is that that all of the companies up here don't just have a crowd. Its a group of people, or community, that specialize in what they do. One very important thing about this type of community is that the crowd needs to be compensated in the right way. You need to over compensate the really good people and not the bad ones. That's the only way you can build your pyramid of community or crowd correctly, which is critical for the business model.

There is one thing I would like to say about corporations using the crowd. Although we launched in 2008, we really started to focus on user acquisition in 2009. Initially, our business plan targeted start-ups, so we went after those companies first. Why? Because they don't have quality assurance, they develop very quickly, its easy to sell to them because they don't need a lot signatures, extensive legal contracts, and need to save money.

However, we already began to see large enterprises in Q1 looking at means to utilize crowdsourcing in different ways. The economy is one of the reasons why we saw this. But obviously, if you look at the market place and the crowd, the business model and price ratio make it very appealing to consumers, start-ups, and enterprises.

Jay Batson:

I would like to take the other side of the equation, and want to preface my remarks by saying that I focus mostly on software. If you're doing something outside of software, then these comments might not apply to you.

In the software open source world, we see two kinds of open source companies. One is where a company tries to commercialize an open source software that already existed. The other is where a company identifies an opportunity and tries to create a business around that using open source as the way to do it. The later is what [Doron] is likely talking about, where you try to create a community, compensate the community, and motivate them to come around the particular thing you're trying to do.

The other type of community is one that does it for their own purposes. An example of that is Linux. No one compensated anybody to anything for Linux. Yet, its the biggest open source success story on the planet.

Up until recently, Red Hat was the only company that made any substantial money in terms of exit valuation of the company. Red Hat made money by working on Linux, which they didn't write. MySQL has been the second now with an exit. MySQL was devoted more around the concept of "we're going to start something." But, it wasn't always that way. MySQL started off as more of a ground sourced project and as they went along they did something structurally to say their going to control what is going to happen.

So, we've had one exit in both areas. But, it doesn't mean that one or the other strategy is the right way to do it.

We're seeing more open source companies today where they identify an opportunity and try to put a company together to go after that opportunity by using open source as a means to get there.

Acquia Drual is one of the fewer exceptions where there was a previous project and there is no motivation of the community by Acquia. The community completely motivates itself, we don't pay anyone, and yet we get one of the best pieces of web technology and software.

So, it really depends in the open source world what model you want to choose.

Peter LaMotte:

The 2 cents that I would add is that the tenants of crowdsourcing company are still the tenants of the company. So, you need to have a simple product in the sense that its easy to understand. For the community, you have to provide community management and the ability to keep them captivated.

There's no secret sauce to crowdsourcing. But, the same way you students are constantly learning in your MBA program, you need to be constantly learning about the community.

For example, we had a competitor that went under who crowdsourced video. They went under because their model contained algorithms that determined incentives based on pageviews and traffic, that were ultimately too complex. We made our platform very simple: submit, select, if you win, then you get money.

Keep it simple and see if there is a niche that has problems the crowd can solve better. If you can do this faster and more affordably with the crowd, then there is probably a good business model behind it.

Question from the crowd:

John, do you compensate your community financially?

John B. Rogers:

We do. In fact, we've run both kinds of competitions on our site - competitions where there is no monetary compensation. But, even those competitions have compensation. I don't know how Jay will come down on me for this...however, when you get known for doing something, that in itself is a form of compensation. If that didn't exist, then you better be careful. If they can't get fame, whatever that is, be it a t-shirt or name recognication, then they're going to stop coming.

However, we do have competitions that we compensate with money. One of our advisors, who I respect as one of the greatest minds in crowdsourcing, Karen McConney, has always told me that its important to figure out pricing early on. There is a wrong price on the high side and there is a wrong price on the low side. If you pay too high or too little, then you'll experience bad behavior.

We've seen this, tested it, and come to a conclusion that he is exactly right. We paid a $10,000 compensation for a car design and got a lot of good designers. But, if they didn't win, they would say "what is this, Local Motors sucks." Then we paid $500 for a car design and didn't get anything. So, we settled for $2,500 mark. However, for different competitions in the future, we may pay more or less depending on what that is.

Question from the crowd:

I bet with Jay's community, you get a small, but very tight group of committers. But, on the monetary side a broader yet lighter commitment from the crowd. How does  incentive design - monetary vs. non-monetary - impact the commitment and intensity you get from the crowd.

Jay Batson:

Well, let me give you some statistics on Drupal. Drupal 6, which is the current release, has 450 active contributors to its core. Just like Linux, there is a central kernal surrounded by a bunch of utilities on the outside. Those utilities on the outside are called "contributed modules" who the larger community contribute as add on modules.

There are already over 900 contributors to Drupal 7, the current release in development. And, that's just the core. There are over 4,000 Drupal contributed models. That number doubles every year. When I first started raising money it was close to 2,000 and the year before that there were about 1,000.

You do earn "Drupal Cred" informally. But our community really isn't defined by 20 or 10 people who have all the "Cred". It's a really broad community and it's pretty easy to get "Drupal Cred" by starting to contribute Drupal patches, modules, or Drupal code. So, its pretty easy to get on the radar screen and get to a point where people know you.

Doron Reuveni:

I completely agree with the two currencies. We have two forms - one where we pay them and the other where they receive recognition. Every cycle that we run has a top tester, who gets paid the most and receives recognition for holding that position. I'm not sure which one they value the most, but to them, it's critical to get recognized as a top tester.

The second thing is that our market place really depends on supply and demand. So, if we're testing a web application where many people can test, then the rates will go down for testers. However, if you look at Blackberry testing in Europe, where there aren't as many testers, the reward is significantly higher. So, we have to make an algorithm that's based on supply and demand to come up with the right rates. It's what drives interest and real value.

Peter LaMotte:

I think we're kind of the odd ball because we have a community of artists who opinionated. And, what you don't want, is a community of 10,000 angry artists.

The non-monetary incentive is something that shows across [different models] well beyond crowdsourcing. If you look at Legos and EA games, and the typse of things people will do to earn "points, it's just amazing to see what people will do for noteriety. But for you us, our people need money. 

Within our community of 10,000, we polled and found out that roughly 67% of the people joined to make money. The next largest percentage is for people who joined and wanted access to types of clients they wouldn't have otherwise. So, it's important for us to set what they feel is a reasonable price.

Question from the crowd:

Is this a new labor model? And, can you possibly make enough money to make a living?

Peter LaMotte:

Our CEO was recently in the New York Times, and recently described it as being a "free agent economy," which I think is a good classification of it - being able to pull specialization from all over.

Doron Reuveni:

To answer that, absolutely yes. But again, you still need to specialize and be good at what you do to be one of the good contributors. If you can do that, then you'll probably be able to make some good money.

Even today, in our early stages, we have testers who are making between $2,000 to $3,000 a month.

Another thing to consider is that, right now, we're thinking about the world as all the same place. We still pay the same amount to testers in India where the cost of living is much less. So, I think it makes the whole world one economy, which is a very unique thing - and participation is increasing.

John B. Rogers:

I don't like the whole labor argument that much. Because even though its inextricably tied to people who are part of crowdsourcing, providing solutions, and are equitable to a labor pool - it's really an unintended consequence of the fact that you're really looking for are great ideas. It's one in the same in the sense that you can't have one without the other. But, to just say that its a new labor model, is not fair.

Because it's like saying "if you don't work with Bill Joy, then where ever you work (since its not Sun Microsystems hiring only the smartest people), the smartest people don't work with you." And, its not that the smartest person doesn't work right next to you. It's when you bring the person into the crowd and gets teased by every person around them, that causes them to become smarter. Especially in very creative elements, that crowds tend to aggregate around.

So...yes, its labor...but, it's ideas that become more intensified and grow to become better within a crowd environment - rather than putting the smartest person to work next to you, and him not be incentivized to do things.

Jay Batson:

In terms of the later model...I think it works in one sense. If the "things" that these people are doing align with what you want to do, then there are advantages that arise both in terms of economics and innovation. If they don't align well, then it doesn't really matter.

But, when you get a community that is passionate and who have some reason of their own to do what they're doing in addition to what your reason is - then it works.

Question from the crowd:

How do you manage the crowd?

John B. Rogers

One of the things that hasn't been said here yet, is that you've now uncovered a new stakeholder in crowdsourcing. You learn in business that stake holders are the ones that can sue if you do wrong, but also be your greatest advocates if you do them right. We usually think of stakeholders as investors. Now in business school, they also teach that stakeholders are also employees. But suddenly, your stakeholders are the crowd, which your customers can also be embedded within. So, it's important to realize that now your stakeholders are part of the crowd, many of whom you're now paying.

So, if you look at Peter's model, they have a very good way of aligning the crowd with what the people want. Projects define the name of the person, what they want, how much their willing to pay, etc. making it very easy for the stakeholder to feel a part of the crowd.

Similarly, you need to be able to manage when they deliver something to you. You pay them money, give them naming rights, etc. and recognizing them as part of the business is a core part of your business in terms of managing them. 

Question from the crowd:

How do you manage or process all of the content that you receive from your community? How do you process what is good or bad?

Doron Reuveni:

We have what we call full time "community manager." Because we look as our community and crowd as our customers. At the end of the day, if we didn't have them working, we wouldn't be able to meet the demand.

A full time community manager should have all kinds of tools to communicate with the community. Such as online chat rooms, online video, a forum system, and the website. Right now we're using an application called Spigit, that lets all of the members share ideas and enable the really important ones come up to the top. For instance, we give them an opportunity to provide feedback on a new logo [for uTest] or how our business model should work.

Peter LaMotte:

For the actual inbound content, we have the ability to push a lot of that back to the client. So, it then relies on technology to make it as easy as possible to decide on batches of logos that are submitted by designers for the customer. However, even with technology, there is always a lot of manual work involved in the final decision. We're always looking over our client's shoulders to help them through that path by providing support or even helping them with final presentations like when we helped Amp'd Energy Drink to present what they thought would be the most viral video to Coke.

Jay Batson:

There's only one comment in the open source world. At the end of the day, at least for the development side of open source software, there is no substitute for a "magical" community lead. Such as Linus Torvalds of Linux, Mark Fleur of JBoss, and Dries Buytaert of Drupal.
 
There's no substitute to have someone who is gifted in managing what the desires of the community are vs. what the desires of community's needs are.

Moderator:

What advice can you give to entrepreneurs in the crowd?

Jay Batson:

If your thinking about a software company, and you're not thinking about using a open source, then you better be thinking about the fact that your target market must be so specialized that only a small number of people would want to develop software for that area. But, if its at all horizontal, meaning that it has appeal to a large number of people, and your not thinking about developing software in open source - then you better be careful, because the old model is completely busted. It's the only model that is going to survive the next 5 to 10 years.

Doron Reuveni:

I completely agree, open source is the way to go. We started developing in open source and even have our website powered by Drupal. Open source is the only way to build software that is scalable and cost effective.

Peter LaMotte:

Be flexible and do it fast. Its only about survival of the fittest, it's about the ability to be flexible and to be able to adapt

John B. Rogers:

The only thing that I can say is to just commit. William Hutchinson Murray, of the Scotish Expedition to climb Mt. Everest, is famed for saying, "we tried 11 times, and everytime we got to a place where didn't know if we could make it back, but we committed, and found something on the otherside that helped us to get further." I think that's been the truest advice anyone has ever given to me.

Everytime I've gotten to a place where I didn't know what to do next, if I committed, then everytime something appeared on the other side that I wouldn't have seen if I didn't go.

If you're going to be an entrepreneur, learn that, and go after it because it will be your salvation afterward.

A passage from Murray's The Scottish Himalayan Expedition (1951)

... but when I said that nothing had been done I erred in one important matter. We had definitely committed ourselves and were halfway out of our ruts. We had put down our passage money— booked a sailing to Bombay. This may sound too simple, but is great in consequence. Until one is committed, there is hesitancy, the chance to draw back, always ineffectiveness. Concerning all acts of initiative (and creation), there is one elementary truth the ignorance of which kills countless ideas and splendid plans: that the moment one definitely commits oneself, then providence moves too. A whole stream of events issues from the decision, raising in one's favor all manner of unforeseen incidents, meetings and material assistance, which no man could have dreamt would have come his way. I learned a deep respect for one of Goethe's couplets:

Whatever you can do or dream you can, begin it.
Boldness has genius, power and magic in it!

If you're interested in similar panel discussions and live in Boston, then feel free to join us and sign up on our mailing list.