Okay, so your company has identified the difference between operational CRM, which (for example) provides frontline employees with cross- and upsell data, or tracks call center productivity, and analytic CRM, which predicts customer behavior, identifies high-value customers and guides the design of well-targeted messages. And you've gone ahead and oriented your customer management efforts toward an analytic bent.
So why isn't the damned return on investment skyrocketing?
There are three reasons, according to a white paper from business intelligence and analytics firm SAS. First, organizations don't combine data from across all relevant departments, and therefore miss important insights. Second, while their metrics may tell them about previous and current customer behavior, reports that provide direction regarding what it will be going forward, why it will be and what to do about it aren't part of their analytics packages. Finally, those firms that do have structures that reveal customer insights don't have the ability to drive activities that result in the highest ROI.
Solving the first conundrum is a hardware and software issue, one which requires a robust data warehouse, a solid hygiene program and an open architecture that allows it to integrate data from new channels. The second and third ones require a bit more heavy lifting from an analytics point of view.
According to SAS, true customer intelligence, which will increase CRM's benefit, is the result of several types of analysis used in tandem. These include: customer profitability; channel usage and profitability; product preferences and profitability; bundling/cross-selling/upselling; customer loyalty/churn; demand forecasting; clickstream; credit risk; fraud detection; market basket; customer segment; and event trigger summaries.
While marketers don't have to use all of these analyses, they should choose the ones that "make it possible to predict future customer behavior rather than merely describing past or present behavior or dividing customers into similar groups", according to a white paper from the Cary, NC-based firm.
So how does an organization use customer intelligence to drive the highest ROI-producing activities? SAS cites Gartner Group, which breaks customer interactions into three types, each yielding a different degree of success.
The first is enterprise-initiated, marketing-department-driven outbound interactions. These produce single-digit success rates, as they are usually designed with the requirements of the organization, rather than the customer, as the top priority.
Second is the customer-triggered, event-driven two-way communication, which often produce a 20% success rate, largely because the appeals are timely and convenient for the customers.
The last is customer-initiated, relationship-driven interactions, which have high success rates because they are timed and customized to suit customer requirements in real time. Gartner claims these programs can yield success rates of 40%.
But highly customized efforts require a greater level of customer intelligence than generic efforts do. SAS recommends that organizations implement the following structures to facilitate this:
*Customer intelligence should flow through the entire organization, rather than being retained in silos of intelligence.
*There should be a process to integrate analytically derived insights back into databases. These files should be dynamic – change should be a constant process, and reflect all new intelligence.
*Reporting should be Web-accessible, with information presented on a level appropriate for each user. Companies should use dashboards, when applicable.
This level of customer intelligence will enable organizations to maximize the value of every customer interaction, and significantly improve the ROI of technology investments, according to SAS.
Network
