Shared publicly  - 
 
Crowdcast's Predictive Modeling Bets on Inside Knowledge

In this blog I'm going to talk about how Crowdcast extracts useful knowledge out of the "information sphere" within a company and helps you predict the future. Crowdcast was recently acquired by Spigit, the leading social innovation platform for enterprise.

When will the product ship? How's the financing going? How's recruiting going? Everyone in business wants status reports. Too often, though, the feedback companies get can be biased by a lack of perspective. Or, frankly, an employee's fear of speaking the truth to upper management.

What Crowdcast does is make collecting critical status updates (predictive modeling) into a kind of betting game, and uses the input from people's insights to analyze possible outcomes. Crowdcast taps into the power of a diverse group of people who have a common set of knowledge but different points of view on the same subject. Employees bet anonymously on answers to questions about the future of their product. It's predictive modeling as a game.

I sat down with +Leslie Fine, Crowdcast's chief scientist, at the offices of Singularity University to learn more about Crowdcast, its vision, and to quench my own curiosity of how well these predictions actually work.

"What we do at Crowdcast," said Leslie, "is take the kinds of things people talk about around the water cooler but aren't in the official plan of record. It's all about delivering collective intelligence inside companies, helping companies with what we call social forecasting, taking the power of internal information -- mostly in teams but also in partners -- and aggregating them to better forecast things such as when are we actually going to ship the product and how good it's going to be and how the market may react to it."

Crowdcast came about because CEO Mat Fogarty, who founded the company five years ago, had seen the need for better predictive modeling. At the time, predictive technologies weren't particularly useful outside of stock trading -- they didn't work for the nuts-and-bolts projects within a company. "He wanted numbers and probabilities," Fine said, "and the answers weren't designed for the average person, the average person who wants to come in and say, 'I think it's going to be delayed, I think it's going to be between X and Y,' and be rewarded if he's right and move on. He was really looking to do something a little more business-centric."

Crowdcast is a leveler of opinion, Fine believes. "Inside of companies what we've discovered is there is a big role for anonymity and engagement and recognition," Fine said. "Those things sound like they wouldn't necessarily go together but if you can give people a safe and anonymous way to talk about what they really know in a fashion that is equally valued to superiors, it's that much more useful. One's opinion is heard, and it means something. It isn't just because you're a VP and your voice is louder than mine as an individual contributor."

Crowdcast "can tell you what's going to happen in the future so you can manage towards it," Fine said. "Inside a company you can actually manage the future if you have a better sense of it. Our best implementation is in places where management or middle management has been enlightened enough to say, 'Okay, my people are giving me a really good view of what the future would look like if I don't take action.'"

Crowdcast customers use a software program to express their thoughts, anonymously, by answering questions, knowing that their predictions will be judged on accuracy. "One of the things that we built into our system is also the ability for people to comment and to vote," Fine said. "Sort of Reddit-style with people's comments, all through a dashboard. That allows people to say, 'I think we're going to ship six weeks late because we have promised supplier X.' He just said that anonymously to management. Other people are voting that not only what's going to happen but why. And then management can come in and either react to that or close the loop and send out a message to the people without identifying them and say, 'Hey, we changed to supplier Y. How do you think that's going to affect the future?'"

For example, Crowdcast worked with a large airline manufacturer that had a problem delivering its innovative new aircraft in a timely manner, which proved costly and embarrassing. "We started working with some people at a very senior level in the company," Fine said, "who believed that while there were technological challenges... there were also informational challenges. Often in companies like this one there are a lot of levels of management, and not a very clear path to go talk to senior management about a challenge you see." In addition, Fine said, "As with so many companies, there can be finger-pointing, in which it's someone else's fault. So we came in and started forecasting the very small milestones that led up to those major deployments." The Crowdcast project started with a few hundred people and has proved so successful it's going companywide.

A large or a small company can use Crowdcast's methods. "It's really not about the size of the company so much as the complexity of the group," Fine said.

Fine outlined three prime motivating factors for people participating in a Crowdcast project:

1. What they say matters; their opinion counts with management. "It's really a very good tool not only to get information but to create a dialog to feel that he is being listened to," Fine said.

2. It's an acceleration tool for change. "You're often well-trained as good managers not to go to your boss with a problem but with a solution," Fine said. "But that solution may not be immediately apparent." Often people play a sort of game of telephone when someone hears something that might be bad, the story gets diluted by the time it gets to the right person's ears, making it essentially meaningless as a measure of a problem to be fixed. Crowdcast eliminates that: it provides accurate assessments of where things are. "It's not so much a crisis as an opportunity," Fine said. "It's a speed tool as well."

3. Prizes and recognition -- and action. "While this is funny money, we do strongly encourage companies to do some kind of giveaway for people just to kind of get the ball rolling and a little excitement going," Fine said. She also recommends "motivating through leaderboards so you can see what your status is when compared to everyone in the organization and in your own team. We usually create teams running a work chart. How's marketing doing versus product, how's product doing versus operations and within operations, who's performing? The prizes kind of get people in the door, the leaderboard entertains people for a little while. Management then has to close the loop and thank people for their input, saying, 'We're going to now do X, Y, Z.' That really gets people."

Crowdcast works to extract useful knowledge out of the information sphere within a company. "The loudest person in the room doesn't always have the information," Fine said. "That's a lot of the value of what Crowdcast does: Finding out who knows what and not just who knows how to speak most loudly."

In my next blog, I'm going to look at an amazing company called Tongal that is revolutionizing the way anyone can create a TV commercial.

NOTE: As always, I would love your help in co-creating BOLD, and will happily acknowledge you as a "contributing author" for your input.  Please share with me (and the community) in the comments below what you specifically found most interesting, what you disagree with and any similar stories or examples that reinforce this blog that I might use as examples in writing BOLD. Thank you!
23
6
Jean B Sawadogo's profile photoAntonio Stark's profile photoFernando Mendirichaga's profile photoNikolaj Berntsen's profile photo
12 comments
 
Thanks Peter, great post! #OpenEnterprise  - it was like that in the early days of the BMW Plant Leipzig. Otherwise quick ramping up production with teams based in various places such as Munich, Regensburg, Dingolfing would have never been possible. #SharingIsGaining  
 
It is strange that many people pay more attention to people and allow more trust to machines. But I reckon it isn't that surprising seeing that people can see problems that machines don't ask, but generate less applicable solutions to existent problems. Are people out of capacity to mold answers? I think what people lack is the coherence and structure which is the medium to solutions, structure which machines have innate.
A good analogy for this which I learned while doing (school) R&E is energy. The solutions are like energy, they're everywhere and abundant in quantity! Nevertheless, humans had to rely mainly on fossil fuel because they didn't know how to utilize and apply the cosmos, and thus turned to scarce, but more readily accesible pools.
I don't know if this is a good enough example, but I feel the nessesity of a structure like Crowdcast when solving mathematical problems with my peers. Many problems, have multiple ways of leading to the answers we need. We(me and my peers) never undermine the importance of an approach, for although it may incur the same answers, the insight it gives is substansially different. It is so even if one approach when thought up by different person. And scarcely did solutions come from a monopolious prodigy, or the start to the end. Everyone had different sections, appliances, which is jigsawed into the whole puzzle only if another person triggers the step below. Solving math seems to be a more tangible way of viewing Crowdcast, a fission reactor of ideas(those who know the exponential growth of U-232 inside a fission reactor exclusive of a stabilizer will understand)
people grow exponentially(or combinatoroly-see +Mindey I. 's post for this) in power, wheras machines do so a degree less. Let us tap into this resource
 
Brilliant idea; I definitely see a use for that channel of feedback with predictive power in the old enterprise corps, especially the one where people on the floor are not really heard.
 
That is really a great Idea. it is a merging of a hierarchical business structure with something similar to DARPA's network challenge, or with the way an open source projects work. 
 
 "The loudest person in the room doesn't always have the information,"

According to studies. In a business environment it is often the first person to speak that ends up having the idea that is followed, right or wrong. 
 
This is an excellent idea, it is somewhat similar to do market surveys... but asking our staff about future changes or the problems we have to solve...  

The great challenge is to be modest and learn to recognize what others can teach us, even people that report us within the organization.
 
This reminds me of the Alan Kay quote: "The best way to predict the future is to invent it."  While obvious for a liner design process, I had never thought of using multiple active and self-adjusting feedback loops to continuously and simultaneously predict and invent the future of an organization.  Very mind-bending.  

[For you Asimov SciFi fans out there, this might be an early predecessor of "psychohistory". :-)  ]
 
This sort of innovation is exactly what motivates employees. By including the actual thoughts and opinions of the employees that are in the know and have useful opinions about how to improve processes, business are allowing themselves to grow in ways that traditional managerial decisions did not allow. Recognition is key in this process, and Leslie mentions it in her third point. By including prizes. Leslie touches on the incentive based competition idea, an idea that Peter pitches continuously. This type of innovation in the workplace is exciting, and I have faith that this is only the beginning.
 
This is a very interesting and extremely important approach for most organizations. When leaders are smart and willing enough to create and nurture a culture of truth telling where everybody is welcome to voice their opinion no matter their rank or capacity for speaking loud, things work out well. This is not the case and I would guess that 99% of organizations might benefit from such programs as Crowdcast! 
 
I like the idea, but I think , the organization must learn to use the tool and to include it in its corporate culture
 
+Nick Fiolitakis you are touching a very relevant issue in our times, "organizational culture". This might be the hurdle, and the accelerator at the same time for rapid positive change adapting to innovations (in any field, ranging from production to finance). #EdgarSchein 's "Organizational Culture and Leadership" http://bit.ly/15ZU4sO gives practical to use insights on what is going on in cultures. cc +Abiel Guerra +David Hawthorne 
Add a comment...