Do marketers really need real-time predictive analytics?

By Wilson Raj, Global Director, SAS Customer Intelligence

The term “real-time predictive analytics” has many dimensions and implications, such as real-time marketing, real-time data integration, real-time customer intelligence, real-time reporting, and so on.

Real-time predictive analytics goes well beyond “real-time web analytics.” So, it’s important to have a grounded view so that businesses can establish a more comprehensive and realistic approach that is suited to their objectives and capabilities.

Real-time analytics is the use of (or the capacity to use) all available customer and enterprise data, processes and technologies to serve customers and the business itself at “the point of need” — or, put another way, “the right time.”

Differentiate between the customer-facing processes and operational marketing processes so that you can appropriately determine how real-time analytics can add tangible value to each aspect.

With the customer-facing processes, businesses can determine how they are executing on their marketing strategies and tactics. Here, real-time analytics essentially allows marketers to create value for their customers. With marketing operations, business can tell how well they are running their marketing and how well they are optimizing all of the available resources at hand — people, budgets, technologies, content assets, etc.). Here, real-time analytics give businesses an instant edge for improving marketing performance and reducing costs.

Real-time customer-facing processes

With customer-facing processes, real-time predictive analytics can be applied to a number of scenarios that create value for customers. The biggest one is real-time offer management where businesses can instantly queue up next best actions.

For instance, if a customer calls to cancel a cable TV subscription, an attrition or risk or even customer profitability score can be calculated with real-time analytics. The service rep can then present pre-emptive offers or take pre-determined actions to induce the customer to stay.

Another example is real-time customer response and support. With real-time social media analytics, businesses can listen, manage, and respond to customer feedback almost instantly. Another customer-facing scenario is real-time decision and interaction management. For instance, marketers can use real-time analytics in campaigns to determine customer eligibility and the likelihood to respond, and then surface a set of relevant offers or next best actions, and then present the best option to the customer.

Real-time marketing operations

With marketing operations processes, “point of need” scenarios can dramatically improve marketing performance and reduce costs. A key example is real-time insight development. As customer data from multiple channels get refreshed, real-time analytics can re-score, re-segment, and re-evaluate customer profitability so that marketers can respond with timely, relevant campaigns and touch-points.

Another example involves real-time digital marketing. For instance, marketers can use real-time bidding (RTB) and analytics to create, buy, and serve digital ads that are customized and dynamic. Yet another example is real-time customer intelligence in terms of real-time visualization where data is instantly turned into useful and actionable analysis instantly — a far cry from the routine, transactional reports that comprise many business dashboards today.

So, how can you ensure that real-time analytics enhances your customer and marketing operations? Here are four tips for applying real-time approaches:

1. Identify and prioritize where you want to implement real-time analytics. Do you want to create customer value through customer-facing processes? Or, do you want to improve marketing performance and efficiencies through marketing operations?
   
2. Assess the cost-benefit of your real-time analytics implementations. For many enterprises, it’s not merely a technology solution. You have to factor in culture implications, infrastructure needs, processes and people.
   
3. Understand the entire ecosystem of your customer and your part in it. Before embarking on a “real-time” marketing and analytics binge, take time to understand the various touch-points in your customers’ journey. Are there partners or other entities involved? What are the interactions you can directly control and impact? What are the hand-off points during customer interactions?
4. Do a self-assessment on the predictive and analytic capabilities you have before embarking on “real-time analytics.” Look at the extent of predictive, scoring, and interaction management models you have in-house and bridge gaps through talent acquisitions, partners, and vendors.
   

The Big Data dilemma shows that sifting through the influx of customer information is a tough task to tackle. When you establish a definition of real-time analysis as it is applied to customer-facing and marketing-facing processes, then companies can take practical, discrete steps to remain relevant in the "real" world of overwhelming data.