It’s hard to believe that more than 25 years have passed since the term “business intelligence” (BI) came of age in the lexicon of IT, business and marketing organizations. Though the fundamental concept of transforming data into knowledge for better, informed decision-making is mostly unchanged, the technology underlying this capability has significantly matured.
With the rise of BI, we’ve seen the information worker go from the exception to the rule. We have also witnessed an evolution from pre-staged data – from data warehouses and hypercubes to real-time data access – that has brought a revolution in Big Data. I’ve said for several years that Big Data on top of next-generation in-memory databases is the new “killer app”, just as the spreadsheet was in the era of personal computers. We no longer think twice when providing our marketing stakeholders with real-time reports and using data-driven dashboards and predictive analytics.
Geeks win – and it’s chic to be one
As a newly minted MBA decades ago, I worked in several roles where I used large data sets and analytical tools to provide analysis, reporting, and recommendations to my clients across both consumer and business-to-business (B2B) industries. Over time, businesses have come to embrace the concept of data science. Like BI, it’s still the extraction of knowledge from data for improved decision-making, but the field has grown to encompass critical ideas from mathematics, statistics, information theory, and IT. Data science has steadily matured into an excellent example of a cross-discipline spanning math, statistics, and computer science as an academic field of study.
Like modern marketers, data scientists are not expected to know everything. However, they should be proficient in a few critical areas and be the tip of the spear to leverage analytics and Big Data. The demand is so high for data scientists that I wish I could jump back in time to recategorize my earlier job titles as a data scientist and restate my previous compensation!
Businesses have embraced marketing performance management, reporting, and analytics as the tools to report progress against specifics key performance indicators and marketing measurements. Many have adopted the SiriusDecisions Waterfall and rolled out an enhanced effort that supports the model with a marketing automation platform (MAP) initiative. My organization, for example, uses a mix of predefined reporting for focus areas and measurement, as well as ad hoc reporting tools with exportable charts and graphs in a variety of formats such as e-mails, presentations, and documents. Performance Management is key for the science part of modern marketing supported by KPIs and measurement.
A relatively recent innovation is a new generation of marketing boardroom tools driving visual, graphics-rich conversations around marketing performance – all enabled by the same underlying and consistent data sources in real time. My organization has an internal group that not only builds these best-in-class tools and reports, but also organizes them in an enterprise analytics store. Borrowing from a concept of the consumerization of IT, this analytics approach is searchable and supported with definitions and documentation in an always-on format for quick delivery. When stored in a cloud-based repository, our reporting assets and updated historical view assist marketing stakeholders in a way that we could have only imagined a few short years ago.
Predictive analytics: The next frontier of analytics, Big Data, and performance management
When it comes to the evolution of predictive analytics, I often use the analogy of driving a car and looking out the front window, rather than as the rear window where all you can see is past performance. One area we’ve embraced is propensity modeling – a technique that detects patterns in data, such as combinations of behaviors and characteristics, to predict future action.
Propensity modeling is a potent tool for identifying and prioritizing buyers. In B2B marketing, the concept of propensity modeling is often used interchangeably with predictive lead scoring. The latter is a crucial concept of marketing automation included in the waterfall that provides better optimization of resources and scalability for actions and follow-up.
For example, borrowing from the Pareto Principle of Economics (also known as the 80/20 rule), we can forecast a relatively small segment of our buyers who represent the majority of our sales. When we use this information in conjunction with other techniques such as relative targeting and segmentation, we gain a more transparent, useful view of not only the addressable market, but also a better way to judge the opportunities within it and our prospective buyers.
So, the next time you hear the statement “the geeks won,” you’ll know why. And you have them to thank as you embrace a future of modern marketing running on marketing analytics, performance management and data science.
Fred is the senior marketing director for SAP HANA Enterprise Cloud and Digital Business Services Marketing at SAP.