We return this week for Part II of our blog with astrophysicist and data scientist, Kirk Borne, Ph.D. Formerly a NASA scientist, he’s one of the foremost experts in big data and its applications in business, government and science − from exploration of space to economic growth. Here again he speaks with Anametrix CEO Pelin Thorogood, this time identifying the areas where he thinks business will benefit most from big data.
Pelin: So, let’s talk about the importance of big data in business. Where do you think data discovery analytics of the type we offer at Anametrix will fit into big data solution sets? Can we talk about implications into how business is done?
Kirk: Yes, I know that Anametrix has a strong solution in this new paradigm of discovery analytics. A traditional dashboard − a business intelligence dashboard − is already pre-programmed to show you certain metrics, certain features in the data. These are metrics that are being tracked on a regular basis. They can be helpful, but basically you’ve already decided what you’re going to look for and find. These are likely to be in the category of known “unknowns.” But the biggest potential in business is to find the unknown “unknowns.” That’s a new paradigm in discovery analytics as you have at Anametrix. We can predefine the metrics, but you can also explore the data in many different dimensions, in multiple correlations and do discoveries that can lead to new areas of knowledge.
Pelin: Yes, the old generations of business intelligence define the questions you are going to ask and ask again in predefined cubes. So you are basically limiting your discovery to the “known.” This is quite different than asking questions of the data and based on the dynamic nature of responses, ask more questions.
Kirk: Exactly. In my early days at NASA, I worked on the Hubble Space Telescope and its data archive. We were proud of that archive because scientists could make any query, not just the ones that were pre-planned and pre-programmed into the interface. That was quite noble at that time. You could scan the top 5 percent of the database, find the most interesting information and drill into it, instead of trying to do the whole all at once. In business these same capabilities enable you to go deeper and ask much more interesting questions, particularly if you can get to those questions faster. So speed is important, given competition. But so is the complexity of the data in volume and multi-dimensionality. I’m excited about being able to ask deeper and deeper questions. The computational capacity is hundreds of times faster today than even 10 years ago. But the volume of data also is growing at the same rate as the computational power. We’re always going to need these ad hoc incremental queries to guide us through the avalanche of data we’re looking at.
Pelin: So what is your list of the top three areas in business that will benefit most from big data?
Kirk: The three I came up with may not be the ones a business manager would, but here’s my take and what I’m hearing from a lot of sources. The first is automation where you are basically looking at ways of improving business decisions. One of the values of analytics applied to big data is that you can learn the decision rules. This is almost like a decision tree model or a business rule model. The model identifies decision points within your data stream. What kinds of actions work best? What kinds of actions are needed under certain circumstances? Let’s say for example, your BI dashboard lets you establish an action to be taken when a metric exceeds a certain threshold. The business process execution can then become more and more automated through the application of the big data analytics.
The second area where business can benefit is cross-channel analytics, the ability to actually discover and engage customers across many different channels. So it’s no longer just the vertical channel. A customer visits my website, and I can track clicks, and then they leave, and that’s all I know about that person. With cross-channel analytics and discovery, you can identify your customer when they visit other sites, either based on some kind of web beacon or user name identity. So I’m a good example. I log into different sites using my Facebook and Twitter accounts. You get to that place where it asks for a user name, password or log in using your social media. So once you do that, then the analytics tools recognize it’s you on those different channels. So cross-channel customer engagement becomes a reality. It builds customer loyalty and actually enables you to discover new customers.
The third one is organizational analytics. That’s a broad term but useful to talk about in specific contexts. One of those is in human resources or talent acquisition and retention. So businesses are looking more and more at the application of multiple data sources to identify top talent and top performers in their company and how to determine the best ways to retain those people. But there are other organizational functions that can benefit from using data, including risk analysis, fraud detection or business optimization. So in these ways the business can use data to improve efficiency, performance and the bottom line.
Pelin: Thank you - great insights about the relationship of data, analytics and business success. We’ll be back with Part III of our conversation with Dr. Borne next week with a look at roadblocks and opportunities in the use of extraordinarily large data sets, including our exploration of space.