November 4, 2013 - FILED UNDER Mindjet
The Science of Innovation: 4 Ways Big Data Informs Innovative Processes
Repeatable innovation isn’t something that just happens on its own — establishing a process to bring the best ideas to market depends on research, data, and the examination of various strategies. In this series, we discuss the science of innovation with our experienced internal Data Scientists, who will walk us through their innovation hypotheses, findings, and unexpected discoveries.
The big data movement has touched every industry and domain. The world is a sea of information, continuing to grow at an astounding rate — in fact, IDC forecasts a 44-fold increase in data volumes between 2009 and 2020.
Analyzing and synthesizing insights from this data, both structured and unstructured, leads to better decision-making, greater operational efficiencies, cost reductions, and growth via new business opportunities. The same is true in the domain of innovation — fully understanding and extracting actionable insights from your innovation program enables adjustments, and ultimately, greater outcomes. At Mindjet, there are several key aspects to your innovation network that we’ve found to be important in achieving great outcomes.
1. It Helps You Tap Your Network
First and foremost, there’s a greater ability to tap into your network. The collective wisdom of a crowd through the aggregation of information in groups has long been recognized as a powerful decision-making approach. Companies that tap into a broader array of employees and customers for innovation tend to generate higher overall collaboration and successful filtering of ideas. And, they have over 55% more collaborative actions per idea in those highly diverse communities than companies who don’t leverage the power of crowdsourcing.
2. You’ll Sustain Engagement
Sustaining engagement around the innovation conversation is vital, and big data can help you do it. Successful innovation requires diversity of thought, relentless consistency, and collaboration. Communities with sustained engagement around innovation produce the best results. Successful communities are those with an engagement profile like you see in the first graph below, with continued peaks. Companies without innovation best practices in place typically have a profile more like the one show in the second graph, with an initial peak followed by flattening.
To help our customers achieve sustained engagement, we have created a challenge-based innovation model, to make focus and activity around innovation natural and easy.
3. Solicit Diversity of Thought
Part of what makes your network so valuable is the diversity of knowledge and opinions contained within it. One of the key ways that opinions and reactions from the network are gathered in collaborative systems is via voting. However, one of the aspects we noticed early on in our innovation networks is that voting behaviors don’t always naturally lend themselves to the best, most accurate aggregated view of crowd preference.
What we’ve found is that typically, when the pool of ideas gets large, the votes per idea follow the power law. That is, most ideas have very few votes, with only a small percentage of the ideas getting a large number of votes. This can be seen in the long tail curve of typical votes per set of ideas, shown below.
And this is why we have created a new way to aggregate crowd preference across a large set of ideas and users, with a voting mechanism called ‘pairwise voting’. In pairwise voting, the system selects the ideas to present to each user in an intelligent way that ensures the pool of ideas gets equal face-time and vote opportunities across the crowd of users.
4. Don’t Forget About Recognition
Another key aspect we see in our innovation network data is that a key behavioral aspect that drives good engagement is social recognition. Those networks that enable social recognition (e.g., user leader boards, reputation, currency, and rewards) do tend to result in networks with qualities and properties that sustain higher engagement.
The properties that we’ve found to indicate that a network is better positioned for higher engagement are: short paths, clustering, and large connected components. All of which develop as a result of the types of social links formed when recognition in the system is explicit.
Big data has a strong role in innovation success. We all need a better understanding of what behaviors, patterns, and trends combine to generate the most optimal outcomes. As we continue to mine our networks for indicators and insights — such as trust, cost/benefit ratios needed for increased collaboration, and overall benchmarks that showcase normal vs. extraordinary ebbs and flows in innovation activity — we’ll unlock myriad truths about our processes, and be able to accurately predict innovation successes for the future.