SAP’s big data processing framework is focused on integrating analytics, big data, and applications in the enterprise. The foundation of SAP’s processing framework is SAP HANA, an in-memory database which improves the performance of analytics applications. SAP’s BusinessObjects Predictive Analysis, the most recent announcement of software built to run on HANA, is designed to help companies identify future opportunities and risks.
Timo Elliott, senior director, strategic marketing at SAP BusinessObjects, readily acknowledges the uncertainty around the term “Big Data.” In a presentation at this years SAPinsider’s BI conference in Las Vegas on Feb. 28, Elliott pointed out the opportunity to gain new insights into big data stores lies in leveraging these new capabilities for business improvement, and said that SAP’s role is to provide “innovation without disruption.” In a follow-up interview conducted by e-mail while Elliott was presenting at conferences in the Nordic region, he discussed his view of the big data trend and how executives should approach these projects.
DI: You tend to be straightforward when you talk about analytics, so how does SAP define big data?
Timo Elliott: “Big Data” means lots of things to lots of people. It was initially associated with open-source technology such as Hadoop, but has since been broadened. I think the most useful definition is “extreme data”, data that is at the far end of one or more of the “3Vs” – volume, velocity, and variety (and organizations shouldn’t forget the fourth V, validity – any analysis is only as good as the underlying data quality). We believe organizations need a “big data processing framework.” It’s not about any one technology; it’s about being able to leverage a seamless selection of different database storage and access mechanisms for particular needs, from real-time complex event processing to in-memory databases to column-store warehouses to massive amounts of data stored in Hadoop. However and wherever data is being stored, you should be able to get its full value through your applications and analytic tools.
DI: In your opinion, do people spend too much time focusing on an explanation of what exactly big data is?
Timo Elliott: I think all “nomenclature wars”, as I call them, are a waste of time (business intelligence vs. business analytics, for example). What’s important is identifying the business opportunities in your organization, and what technology can be used to deliver on those opportunities. If you can do that, I don’t think anybody will care what you call it.
DI: During your keynote at SAPinsider’s BI 2012, you mentioned that the role for SAP customers is to take big data and business intelligence and turn it into business change. Could you provide a couple of examples?
Timo Elliott: You can read about many of them in detail on the experiencesaphana.com web site. They would include organizations like Colgate Palmolive and Provimi. They had amassed huge amounts of detailed information about their businesses, but their existing infrastructure simply couldn’t provide insights in a reasonable time frame until they invested in SAP HANA’s in-memory technology. And T-Mobile is a good example of using the new possibilities to not only “improve analytics”, but to change the business models in the industry: it is now able to offer customized rate-plans on an individual basis, rather than one-size-fits-all segments: their “products” couldn’t exist without the new analytic technologies.
DI: And if someone says “I want to be a champion of that business change,” what are the first two or three steps they have to take?
Timo Elliott: First, congratulations! I personally can’t think of a better background for leading the companies of the future than a detailed understanding of the digital heart of today’s organizations.
Step 1: Work on understanding key end-to-end business processes, and how people are using (or not using) information today.
Step 2: Identify the key “decision points”, and lead “what if” discussions with business leaders. For example, Fresh Direct was able to make big changes to customer satisfaction by asking themselves the question: “What if we could predict in advance if our deliveries were likely to be late?”
Step 3: Provide something tangible for business people to see and use. Many business people find it hard to understand the new possibilities that IT people take for granted, and so they ask for slightly better versions of what they already have. I’ve found that prototype dashboards and exploration views, using fake-but-realistic data, are a powerful mechanism for changing the conversation to “what could be.”
DI: Are companies that don’t use big data and analytics at a competitive disadvantage?
Timo Elliott: We now have the possibility to unlock the value in unstructured, web, and social data, and there are soaring numbers of mobile devices and sensors creating massive amounts of new data (the so- called “Internet of things”). It’s hard to think of an industry that won’t be transformed by the new possibilities, and I think most people understand that. I may be overly optimistic, but I think that almost a decade after The Innovator’s Dilemma was published, and after seeing a number of well-entrenched companies flounder and startups become titans, I’d like to think that executives realize that their organizations can’t afford to be complacent. Our job at SAP is to make it easy to take advantage of the new technology, through “innovation without disruption”, such as SAP Business Warehouse on HANA or the new packaged analytic applications such as Smart Meters for Utilities.
DI: You encourage companies to pull away from being reactive and try to become more proactive by thinking about business analytics in real time. What do you see as your customers’ greatest technical and cultural challenges to taking a more proactive approach?
Timo Elliott: There are two main problems. The first is inertia and vested interests: in many organizations, it can actually be very hard to do things in simpler ways. As author Upton Sinclair once wrote: “It is difficult to get a man to understand something when his salary depends upon his not understanding it.” The second is imagination – it’s not about the quality of the tools: giving somebody a pencil does not make them Picasso. The increasing power of the technology available has meant there is even stronger demand for professionals who can translate that power into business success: a recent survey by LinkedIn (using big data techniques!) showed that demand for “data scientists”, “business analysts” and similar job titles was skyrocketing.
DI: A recent blog by Mark McDonald of Gartner stated that traditional “BI is dead”. Is big data driving the path for real-time business analytics? Do you see a time where we rely exclusively on real-time business analytics? How would that compare to traditional BI today?
Timo Elliott: To be fair, he said “without the business in business intelligence, BI is dead,” and I completely agree. Business intelligence, or business analytics, or whatever you want to call it, is not only alive and well, it’s growing at over 16 percent, according to Gartner, faster than just about any other segment of enterprise IT. The new technology possibilities just extend what it’s always been about: helping organizations make the best use of their information assets. What big data (and other related technologies, like predictive, mobile, cloud and collaboration) does is extend the possibilities to new business areas.
DI: How has big data changed CIOs outlook on data quality and security?
Timo Elliott: I’d like to think that it’s helped accelerate the existing trend towards better information governance. Every industry survey continues to show that the biggest barriers to getting value out of business information are problems with data integration and data quality. These are not IT issues, they are business issues: IT and business leaders have to collaborate to improve the underlying business processes. One of our hottest product areas right now is enterprise information management, with new products such as SAP BusinessObjects Information Steward that uses analytics to improve analytics, letting business people and IT collaborate around data definitions and track data quality key performance indicators.