Radical Innovation And Machine Learning

Savannah Voll

Innovation may be the calling card of the future, but how can we determine that we’re transforming in the right way? Namely, how can we sustain innovation when the digital pace of change increases exponentially?

It begs the question: Are all the good ideas taken?

Innovation experts broke down these questions and more in an episode of the Game-Changing Predictive Machine Learning Radio series called The Machine Era: Disruptive Innovation and Ideation. Host Bonnie D. Graham led three panelists through a discussion focused on how we should innovate in the age of machine learning.

The featured experts were:

  • Bryan Mattimore: Co-founder and chief idea guy of The Growth Engine Company and Innovation Agency
  • Omar Maher: Practice lead for machine learning at ESRI and co-founder of two tech startups
  • John Schitka: Senior director (retired) of solution marketing at SAP

Check out the full replay online. Here are the highlights.

What holds us back from innovation?

Is it the tendency to get caught up in everyday tasks that keeps us from generating good ideas, or does today’s innovation overload keep us from even trying? The panelists dove into a debate about the biggest innovation inhibitor: Is it complacency or saturation?

Complacency of technique

Bryan believes it’s complacency, but added that it’s more about the method than the habit. While there are hundreds of new ideation techniques designed to help us innovate, we often fall back on the tried-and-true team brainstorming meetings. The problem is that these are no longer working.

“You have to introduce stimuli to people’s brains to get them to trigger new ideas,” he said. Open brainstorming sessions “may be okay in the beginning of a project, but, boy, as you move down the road, you’ve got to come up with new ways to trigger people to get new ideas.”

Fear of the unknown

John believes it’s neither complacency nor saturation and instead has more to do with comfort and fear of the unknown. “I’d prefer to go down a path that I know and that I’m comfortable with,” he said, “then walk into a dark room and not know what my feet are going to hit as I step forward.”

Lack of understanding

Omar agreed with both Bryan and John, but pointed out that “another factor is getting to know those new tools in the first place.” Many people hear about artificial intelligence, machine learning, and other hot-button technology topics in the news, but don’t really understand them. That’s the first major creativity roadblock before innovation can even begin. He noted, “One thing I really enjoy the most is demystifying those concepts – explaining them in simple English.”

How can we be intentional about innovation?

Einstein famously said that if he had an hour to work on a problem, he’d spend 55 minutes thinking about it and five minutes solving it. He believed that a well-stated problem is a problem well-solved. John, Omar, and Bryan investigated this approach to problem-solving, also known as the 80/20 rule, and discussed how it can lead to purposeful innovation.

Define the problem

John, a proponent of Einstein’s rule, recommended taking the time to define the problem before executing the solution. One of the biggest issues at the beginning of the innovation process is that having a vague idea of the final step. “What is the problem you’re actually trying to solve?” he asked. “Is it maybe multiple problems, and we’re going to have to break it up into smaller bite-sized pieces?” That’s where we have to start before we can get to creative innovation.

He likened this to using machine learning, saying that all data must be prepared and refined before it’s ready to be used in any kind of solution. “All the work and effort have to go into clearly defining a problem, so I understand what it is I’m trying to solve.”

Confirm the direction

Omar suggested a two-pronged approach. First, we must think about the business challenge as the primary piece of the puzzle. “Think about the problem from a business perspective rather than jumping to technology, which I find many people doing.” Breaking up the question, as John mentioned, is the next step to determine sub-problems. But the most critical piece is confirming the direction of the project. “I think finding the right question is the most important step in all of this.”

Consider the assumptions

Bryan advocates a calculated approach to innovation to garner measured results. One of the most effective ways to start the innovation process is to note all the assumptions about your problem. “List maybe 20 of those,” Bryan recommended. “What’s interesting is that you get the obvious assumptions at the beginning, and then it gets harder and harder, and then you get some real breakthroughs towards the end of that process. That’s why we push people to get 20.”

Then break down the reasoning behind each assumption and determine whether it’s true. For example, a hotel business might assume that it needs to acquire rooms to rent. However, businesses like Airbnb moved past that assumption and disrupted the entire market.

“This questioning of the basic assumptions is critical to the work we do,” he said. Sometimes the key to innovation is reframing our problems.

How can machine learning help us innovate?

Bonnie ended the show by asking the experts how we can use machine learning to innovate, especially in the future.

Data cleansing

As John mentioned earlier, it takes a lot of work to prepare data upfront before feeding it to machine learning technology. However, Omar predicted that it will be common to have data-cleansing programs that prepare the data with little or no human intervention. “There are techniques to do that today, and I think they are going to get more useful.”

Automated prompts

Bryan envisions a future that incorporates machine learning into the innovation process from the start by harnessing its power for question development. “Using predictive machine learning to generate some of those intriguing questions, they could be funneled to the global virtual ideation teams to come up with ideas, almost on a daily basis.”

He ended by asking: “How much more productive could we all be if we had this culture of curiosity?” By blending machine-learning question-generation with human creativity, we’re bound to increase our capacity for innovation.

Continual improvement

John noted that right now machine learning is a great tool, but it will continue to improve as the world changes. It’s been created to solve problems, but the creativity involved is still attached to the human way of thinking. “The ideation, the application of the tool, the coming up with the problems,” he remarked, “will still be very much human.”

Learn more

For more machine learning and innovation thought leadership, listen to the full replay online and check out another episode of the Coffee Break with Game-Changers series. Read these articles to learn more about machine learning.

And please listen to the replay of our “Pathways to the Intelligent Enterprise” Webinar, featuring Phil Carter, chief analyst at IDC, and SAP’s Dan Kearnan and Ginger Gatling.

This article originally appeared on the SAP Analytics blog and is republished by permission.


Savannah Voll

About Savannah Voll

Savannah is a product marketing intern at SAP and is in the last year of her undergraduate degree at the University of Waterloo in Ontario, Canada, where she studies social sciences and business. She loves to write, and specializes in translating technical ideas into engaging content.