How do you analyze innovation at a large multi-national corporation? I’ve been using the Data Analytics Life Cycle (depicted below):
These steps have helped me to internally construct a strategy for analyzing global innovation processes and methodologies at my own corporation (EMC). I’ve published some of the results of this effort in previous posts. The analytic lifecycle was something I learned while attending the new Data Science and Big Data Analytics course. Who came up with these steps? They are essentially an overview of industry best practices and experiences as summarized by EMC Education Services (for a deeper dive into the course content, register and take a look here). The business driver that caused me to attend the course was my belief that analytics could help me discern promising new opportunities in my role as Director of the EMC Innovation Network. The first thing I learned at the course is that I am not a data scientist. Any successful analytics project has a set of key players. In addition to data scientists, key roles include project sponsors and managers, DBAs, data engineers, and business intelligence analysts. My role is essentially that of a business user: someone who consults and advises on how to operationalize the end result of the analytic exercise. The Discovery Phase Step number 1 in the analytics lifecycle is all about the business domain. Here are some of the key activities that are critical in this phase:
- Frame the business problem as an analytic challenge that can be solved in phases.
- Understand what’s been done in the past.
- Assess the resources supporting the project (people, technology, time, and data).
- Form initial hypotheses.
- Determine readiness to move to the next phase.
Framing the Business Problem My company has 50,000+ globally distributed employees, many of whom innovate on a daily basis. The main business problem, from an innovation standpoint, is to ensure that we have an innovation pipeline that continually introduces new revenue sources and cost improvements. Innovation is the lifeblood of the company, especially in the field of high-tech. The most important word in EMC’s innovation lexicon is knowledge. The mission of our EMC Innovation Network is to (a) expand knowledge locally, (b) transfer it globally, and (c) leverage it strategically. The continual introduction of new revenue sources and cost improvements comes down to leveraging new knowledge that has been transferred and shared across our global employee base. Our company needs to analyze the expansion, transfer, and leverage of knowledge. The insight gained from this process will improve the innovation pipeline (one of the hypotheses that I will expand upon in future posts). Understanding What’s Been Done in the Past When it comes to a repository for innovation data, EMC has a five-year history of global ideas submitted by employees. The repository amounts to roughly 6000 ideas. In addition to the idea repository, each business unit has their own repository and collaboration site describing innovation and research activities specific to their business. As I mentioned in a previous post, the idea of running analytics across this type of data is a fairly novel approach. We’ve already analyzed year-over-year idea submission totals, with an emphasis of the geographic location of the submitter. Other than that, there has been no previous attempt to analyze knowledge expansion and transfer on a global scale. This realization surfaces a clear pain point: any future analysis would require some sort of centralization of global knowledge activities related to innovation. This meant that our project would likely be phased. The team would start with the idea repository and focus on measuring research and innovation activities in a later phase. Assessing the Resources The resources required to run this analytics initiative is a good news/bad news situation. The good news is that I work in the CTO Office and my team has an excellent lab with excellent lab managers. Compute/network/storage resources are not an issue. The bad news is that I need two teams of people to help with this project and none of them report into my organization:
- I need sponsoring organizations, managers, and high-level DBAs to help with the initial phases.
- I need data engineers and scientists to execute the low-level work.
I’ve solved this problem by forming two global, volunteer teams within my own company. I met with MIT Professor Peter Gloor to ask his advice on how to motivate globally distributed teams outside of my functional organization. He gave me great advice that worked. Every other Tuesday morning I meet with Team #1: the steering committee for the project. On alternating Tuesdays I meet with Team #2: data scientists-in-training! These teams are distributed throughout the U.S., China, India, Israel, Egypt, Russia, and Europe. Initial Hypotheses and The Analytic Plan This next sentence is essentially the jewel of the course: The hypotheses and analytic plan form the foundation of everything that comes after it. Before taking the course, my hypotheses fell into two buckets:
- Descriptive analytics of what is currently happening in my organization will spark further creativity, collaboration, and asset generation.
- Predictive analytics will advise executive management of where it should be investing next.
After taking the course, I realized that I need to spend much more time a more comprehensive set of hypotheses and a formalized analytic plan. I will be diving into each one of these in future posts. image credit: stevetodd.com Don’t miss an article (4,000+) – Subscribe to our RSS feed and join our Innovation Excellence group!
Steve Todd is Director at EMC Innovation Network, and a high-tech inventor and book author “Innovate With Global Influence“. An EMC Intrapreneur with over 180 patent applications and billions in product revenue, he writes about innovation on his personal blog, the Information Playground. Twitter: @SteveTodd