In a magazine article I recently read about astronomy and cosmology I was struck by the similarities between analytics practitioners, some referenced to as “data scientists,” and multi-wavelength astronomers who use spectroscopy instruments. I enjoy astronomy. How can one look up at the stars and the universe and not exclaim “Wow!”?
Of course, you may be asking, “What is a multi-wavelength astronomer?” These are researchers who analyze mountains of data about stars and star formation in our universe. Astronomical spectroscopy can be divided into three bands across the broad light spectrum: optical, radio, and X-ray.
X-ray astronomy is used to view galaxies where young stars are being formed from the gravity collapse of gases. The resulting radiation emitted is more detectable as infrared, visible, and ultraviolet light across the broad wavelength spectrum. This is light that can be viewed from other parts of the light spectrum. X-ray emission is relatively weak. Hence astronomers depend on multi-wavelength data from all three light spectrum bands for more complete data to place those measurements into context for learning answers to questions they are interested in.
What does astronomy have to do with business analytics?
The magazine article’s point was that multi-wavelength astronomy combines the best available data from every band of the light spectrum for the same object, like a star or galaxy. This is similar to task of practitioners in the analytics and enterprise performance management (EPM) communities. An organization cannot make better decisions and improve its performance by focusing on only one variable, such as only cost, time, quality, service-level and so on. These factors are interdependent. So, it is a much more complex problem. Plus there is much more volatility today, much caused by reduced trade barriers from globalization, which has increased uncertainty about the future. Analysts are on a mission to reduce uncertainty.
As examples of types of analysts, the cost analyst may examine product and standard service-line costs and profit margins. The customer relationship management analyst may examine customer satisfaction and loyalty data. The six sigma quality team may examine quality and root causes of problems. The lean management team may study process cycle-times and throughput rates. The operations analyst may examine production schedules, inventories, and resource capacity levels. However, all of things are connected!
An analyst examines data for investigation and discovery. In all organizations there are a lot of moving parts like gears in a machine. Hence, there are many interdependencies loaded with variables. This is what excites analysts. They have the keys using software tools to unlock insight and foresight from the data. A challenge for them is to determine which keys will work.
The enterprise performance management (EPM) practitioner examines how all of these factors fit together. Their objective is to help their organization perform better, faster and cheaper. But there is more to just those three aspirations. There is also a need to be safer by integrating findings from the enterprise risk management (ERM) community. In addition, all of these factors must align with constantly changing strategic objectives defined by the executive team. Using strategy maps and key performance indicators (KPIs) displayed in dashboards a fifth aspiration is added – to be smarter.
Analytics software provides the technology instruments for research
Business intelligence (BI) and business analytics software provide the instrumentation for analysts and managers. This type of software is similar to what telescopes provide for astronomers – the power to see, to test hypothesis, to understand, and to know.
Each of those types of analysts earlier mentioned are trained in their field on how to think. They learn about techniques like regression and correlation analysis. In contrast, the business analytics and enterprise performance management disciplines, like the multi-wavelength astronomers, train people where to think. Performing good everywhere, the typical “excellence” message that motivational speakers often promote, is important. But it is too simplistic. It may make you feel good to hear about being “excellent” but it rarely lasts. Why? Because there typically insufficient sustaining actions. Focus is required. Focus is what business analytics and enterprise performance management (EPM) methods bring. The objective for any organization is to be better, faster, cheaper and also safer and smarter.
Should organizations be smart or healthy?
A problem all organizations suffer from is their imbalance from how much emphasis they should place on being smart rather than being healthy. Most organizations over-emphasize trying to be smart by hiring MBAs and management consultants with a quest to achieve a run-it-by-the-numbers management style. These types of organizations miss the relevance of how important is to also be healthy – assuring that employee morale is high and employee turnover is low. They miss the need to also assure that managers and employees are deeply involved in understanding the leadership team’s strategic intent and direction setting. Healthy behavior improves the likelihood of employee buy-in and commitment to actions to improve performance – their own and their organization’s performance.
What is needed to correct this imbalance between being smart or healthy? Right from the start, you have to think like a sociologist, and arguably you need to be a psychologist too. People matter – a lot. Strong leadership is needed. Never underestimate the magnitude of resistance to change. It is natural for people to prefer the status quo.
Today the best leaders are not the ones with best answers typically based on their experience or intuition. The best leaders are the ones with the best questions. They motivate their organization to learn.
Like the multi-wavelength astronomers, those organizations with a culture of analytics and burning desire to improve its performance will leverage analytics to solve problems and pursue opportunities.
Try deep learning using MATLAB