When it comes to predictive analytics and modeling, Big Data should be less about grandness and variety and more about practicality. In other words, the amount of data you have on your customers’ actions, inactions and affinities is less essential than the usability of said data.
According to an IBM study published in October, 67% of IT and business professionals worldwide indicated that predictive modeling was among their company’s analytic capabilities. Unfortunately for marketers, Big Data and predictive analytics aren’t the omniscience-bearing technologies we’d hoped for. Consumers have steadily increased time spent with various media types and devices, and while the resulting behavioral, transactional and social information is a gold mine for marketers, it’s all for naught unless they’re able to make sense of it.
Turning information into actionable insight remains one of the industry’s highest hurdles, with the proliferation of new marketing channels and platforms making for a tricky attribution-to-action scenario. At least 82% of US brand marketers and agencies are concerned with their ability to integrate cross-channel data—and 96% aren’t completely satisfied with their ability to understand and drive ROI from big data—leading me to believe that marketers need to refine their approach to procure the most relevant and valuable information.
“What” is not as important as “Why”
Marketers’ superficial application of predictive analytics for personalization is reminiscent of the features found on Amazon and Netflix, using customer data to influence their behavior and advance them toward a specific action like making a purchase, downloading a white paper, or filling out a form. The beauty of massive customer data sets is that with enough historical information, predicting a user’s future action propensity becomes much easier. Relational databases that store and manage petabytes of data can tell us what our customers like and, consequently, what else they might like, but not why our customers like certain products or brands.
While digital disruption fragments attribution and plays a role in disjointing our view of the buyer’s journey, it also brings us closer to discovering the manner in which buyers’ innate beliefs and preferences—established long before initial contact with your company or product—affect their buying paths and choices today and beyond.
Integration with the digital world is underway; we no longer exist in a purely physical world. Similarly, our customers live online, on their devices, across social platforms and in CRM systems, providing data junkies with a deluge of information. By tying customer retention and affinity metrics, as well as social media responses and surveys into contextual, cross-channel attribution data, marketers can gain a deeper (and broader) understanding of how their customers’ thoughts and feelings—not only their actions—inform future preferences and decisions.
Profitability > Growth
Despite the availability of cross-channel campaign management platforms, only 46% of executives agree that their employees have the skills needed to manage digital disruption. While this could point to organizations actually lacking competent employees, it’s more likely that:
- Rapid channel growth has funneled marketers into channel-specific silos (search, display, social, mobile, etc.), leading them to develop cross-channel myopia
- Employees lack the resources or training necessary to follow the organization’s executive-prescribed direction
Adopting a “big picture” outlook on digital disruption calls for marketers to break out of their silos, but it also demands that the flow of information extends throughout the organizational hierarchy, especially to customer-facing positions. Because of their level of contact with customers, sales professionals and technical staff are a premium source of client data—they know why a customer bought and what they’ll need to continue buying.
Realizing the true value of predictive data relies on collaboration between sales and marketing: combining customers’ stated needs with their digital personas, helping to pinpoint the steps in their buying journey most likely to indicate need fulfillment—it could’ve been a white paper, a webinar, or even a LinkedIn message that brought them over the line.
Whatever the case, predictive analytics allow organizations to focus on long-term profitability, rather than short-term growth; instead of aiming to simply sell another unit, a company-wide devotion to creating and nurturing lifetime customers will spur innovation, causing marketers to develop campaigns that instill lasting value in the minds of prospects. Instead of asking: What next touchpoint would make this prospect most likely to buy, based on their previous actions?, ask: What have similar prospects said about us, and what ideal combination of touchpoints will provide actual value and be most likely to meet their predicted needs?
Marketers using predictive modeling in personalization-only mode, without paying heed to their audience’s non-digital human motivations, will struggle to keep up with the customer-centric, data-driven innovation leaders in the industry. If your organization can identify its highest-value customers, uncover customer-specified values and limitations and share those lessons with organizational stakeholders, all while applying them to predictive marketing strategies, you’ll ultimately be able to engage prospects in an optimal digital conversation. Enough datamongering—it’s time to find out what all your data is really worth.