Predictions for the Coming Year: Big Data Analytics

Stephen DeAngelis

December 20, 2013

Pundits have been predicting that Big Data is going to be Big Business for a few years now. Lisbeth McNabb believes that 2014 may finally be the year when companies truly grasp its potential. She notes that there are three challenges associated with Big Data that have hindered its full utilization. They are: “Determining how to get value from data, defining the big data strategy, and obtaining the skills and capabilities needed to make sense of it in a meaningful way.” [“How 2014 Will Be The Year To Monetize Big Data,” Entrepreneur, 13 December 2013] She believes that most of those challenges have been addressed so she predicts, “We are on the cusp of the second wave of the big data revolution. If the first chapter was about investing in the technology to harness the insight and analysis from data, the second is about using big data to quickly get to revenue.” McNabb is not the only analyst thinking about what lies ahead for Big Data. Timo Elliott and Tim Lohman each offer their list of Big Data Trends to look for this coming year as well. Lohman writes, “Throughout 2013, big data and analytics have been among the most-hyped themes within the world of IT. With the vast datasets that most businesses now own about their customers and themselves, it’s no wonder that applying analytics to this data to better understand customers and business processes is increasingly attractive.” [“Big data and analytics trends for 2014,” ZDNet, 16 December 2013] Elliott actually penned his list early in 2013, but he agrees with Lohman about the importance of Big Data Analytics making it number one on his list of trends. [“The Top 10 Trends In Analytics 2013,” Business Analytics, 22 April 2013] He wrote:

Analytics And Business Intelligence Are Still #1. According to Gartner’s latest CIO survey, the top business priority is back to enterprise growth, and analytics and business intelligence remains the number one technology priority for 2013. And the next three technologies on the priority list (mobile, cloud, and collaboration) are all key areas for analytic innovation.”

Alec Gardner, industry consulting director at Teradata Australia and New Zealand, told Lohman, “[In 2013], more companies will roll up their sleeves and ‘have a try’ by starting to bring in new data sources and technologies to answer new business questions and gain insight either not possible or too difficult previously.” The types of projects that companies are investigating include more efficient marketing strategies and better understanding of the consumer digital path to purchase. Gardner continues, “Projects will also enable clearer visibility of behavioural trends and patterns to see both opportunity to enhance a relationship — for example to upsell and improve service delivery — or to intervene to reduce fraud or risk in a business.” To learn more about the latter topic, read my post entitled Fraud and Theft in the Supply Chain. Gardner agrees with McNabb that 2014 could be a watershed year. He told Lohman, “In 2014, we should start to see and hear some significant results from organisations that have been quietly investing in advanced (big data) analytics in 2013.” Gardner’s view also corresponds nicely the second trend identified by Elliot, namely, that analytics has rapidly matured. He explains:

Increasing Analytic Maturity. Thanks to greater industry maturity and new technology opportunities, most organizations are making steps from Descriptive Analytics (‘what happened?’) and Diagnostic Analytics (‘why did it happen?’) towards Predictive Analytics (‘what will happen’) – with Prescriptive Analytics (‘“how can we make it happen’) as the next frontier.”

I agree with Elliott that predictive analytics will become much more important in the years ahead and I don’t believe it will be too difficult to make the subsequent step to prescriptive analytics. Artificial intelligence systems, like the Enterra Solutions® Cognitive Reasoning Platform™, will help speed this transformation. Lora Cecere, founder of Supply Chain Insights, recommends that companies “experiment with cognitive learning engines” like Enterra’s. [“New Products: More Costly and More Important,” Forbes, 11 December 2013] The next trend identified by Elliott involves in-memory computing. I should note that Elliott is an SAP employee. He writes:

In-Memory is Ripping Up The Old Rules. In-memory computing is providing an opportunity to rethink information systems from scratch. According to Gartner, in-memory: ‘isn’t only about SAP HANA, isn’t new, isn’t unproven, isn’t only about big companies, and isn’t only about analytics’: ‘In-memory computing will have a long-term disruptive impact by radically changing users’ expectations, application design principles, and vendor’s strategy‘.”

Enterra’s Cognitive Reasoning Platform can enhance in-memory computing to accomplish some pretty amazing things. It’s because of capabilities like cognitive computing that Elliott’s next trend builds on his previous one to assert that old analytic barriers are coming down. He explains:

Breaking Down Old Barriers. In-memory breaks down long-standing analytics barriers. For example, in-memory computing platform SAP HANA supports structured and unstructured data in a single system, and includes a sophisticated, embedded text analysis engine. Predictive or advanced analytics no longer requires a separate system – powerful analytic algorithms are available directly in-memory, without any unnecessary data movement, and thousands of times faster than disk-based predictive system. Since detailed row data is stored without any aggregation, it makes it much easier to let business people upload their own data sets to the corporate hub for analysis.”

Speed is one of the big benefits that in-memory computing brings to the table. This is especially important since most analysts agree that supply chains are moving towards real-time operations. That fact segues nicely into Elliott’s next identified trend.

Operations and Analytics Are No Longer Separate. For forty years, operational systems and analytic systems have been separate because of technology limitations. That’s now changing with in-memory platforms. For example, with SAP Business Suite on HANA, transactional data is written directly to memory, where it is instantly available without any of the analytic compromises that have plagued earlier ‘real-time’ analytics.”

The fact that Big Data platforms can help integrate data is important. For too long data has been siloed in business units and only reluctantly shared throughout organizations. Brock Douglas, business analytics leader at IBM, told Lohman, “To date, organisations have predominately been doing simple, siloed analytics projects. Moving forward, the culture within organisations will change from a siloed ownership of data and insights to a cross-enterprise approach. Data and the insights uncovered will be shared and applied across the organisation. Data and analytics will no longer be the sole responsibility of the IT function. Who within the organisation data analytics is being used by will evolve. To achieve this, marketing, sales, HR, finance, and supply chain will share the insights gained from data to inform a more holistic business strategy that delivers growth to the bottom line.” McNabb asserts, “From the growth-stage startups to the largest of retailers, we should expect 2014 to be the year that business leaders across the board are using big data to drive results.” That’s why Elliott’s next trend proclaims Big Data a big deal.

Big Data is a Big Deal. In addition to traditional ‘transaction data’, it’s now feasible to analyze ‘interaction data’ (events before, after, and around a transaction, such as the products that were considered but then not purchased) and ‘observation data’ (such as data streamed from sensors). Algorithms such as MapReduce and projects such as Hadoop have introduced new opportunities for storing and analyzing data that was previously ignored because of technology limitations. Actuaries are finding new careers and glory as ‘data scientists’. These new technologies have more than proved their worth in niche or standalone systems, but need to better integrated with existing corporate environments.”

Gardner would probably agree with that assessment. He told Lohman, “IT departments will need to create a unified data architecture that enables business users to access the data and increment their existing models or drive new questions. Organisation cultures may also need to change, moving from a view of analytics discovery as a discrete project to a core competency for the business. In a similar vein, those with the tools to access analytics will need to encompass the whole business, not just the IT department.” Analytics are not just the battleground for the IT department; for that reason, Elliott’s next trend notes that analytics are becoming a core business capability.

Analytics Moves To The Core. Analytics is no longer an afterthought to your transaction systems — it’s the heart of your future information infrastructure. The data you are storing now you will still have in 15 or 20 years’ time, while your applications may be long gone. The next generation of information infrastructures will combine big data, transactional data, analytic data and ‘content’ into a single, coherent set of services that Gartner calls an ‘information capabilities framework’.”

Elliott’s next trend involves ease of use. Numerous articles over the past couple of years have touted the importance (and shortage) of data scientists. More recently, however, it has become apparent that many of the data scientist’s capabilities can actually be built into analytic systems so that non-data scientists can more easily obtain the insights they are seeking. Elliott explains:

Optimizing the User Experience. Today’s information consumers demand the same ease-of-use and immediate access they get in the consumer world. Business people want to be able to grab and mix information on the fly, without having to wait for it to be loaded into a corporate data warehouse. … And of course, people expect a smooth, mobile-ready BI experience with integrated social collaboration, and the option of using a cloud-based infrastructure.”

McNabb insists, “Every single business should leverage the power of data analytics and predictive analysis.” Elliott goes even further. He believes that companies should explore how they can incorporate the collection and analysis of information into all of their products and services to make consumers’ digital path to purchase a good experience.

Information as an Asset. Along with all the technology changes, there have been big changes to analytics culture. Information is no longer a byproduct of manufacturing processes – it is fast-becoming a key part of the products themselves. Today’s retailers and service providers want to offer ‘customer experiences’ that are tailored to individuals, optimized for the moment, and coherent over time – and that requires powerful new data platforms. As information becomes part of revenue generation, interest in information and control over budgets are swiftly moving to the business units, rather than traditional IT. This is creating new opportunities, but also new IT pressures and organizational issues.”

Finally, Elliott notes that as the mountains of data being collected continue to grow, so will the challenges associated with analyzing it. Integration will become an especially tricky challenge. He explains:

The Revenge of Information Governance. As the technology gets more and more powerful, it becomes even more important to fix one of the oldest and least tractable barriers to successful BI: the pain of integrating multiple sets of quality data. Better integration between ‘big data,’ traditional analytic systems, and transaction systems must also involve investments in data governance and solutions such as SAP Information Steward.”

Lohman believes that 2014 will be the year when Big Data is unleashed. There are numerous reasons for this assessment including the increased penetration of mobile technologies, rapidly improving analytical methods, and a growing library of use cases showing how Big Data analytics can provide useful insights that significantly improve decision making.