7 Strategies for Profiting From Customer Data
Nobody knows more about your customers than the people who speak with them every day, whether taking in praise and orders, or resolving criticism and defects. Thus, call centers are uniquely powerful and deep sources of information about customers--how they behave, what they react to, and what truly matters most in your business relationships. These seven strategies illustrate some of the best ways to apply the information you may already have in your contact center to improve your standing with customers.
Know What's Important
Just as you would not use a fire hose to douse a match, you cannot expect your contact center to collect and reflect all that could be knowable about your customers. Choosing the right mix of customer information to track and analyze goes hand in hand with making the right choices about product offerings--as office supply manufacturer and distributor General Binding Corporation (GBC) learned recently. As part of a customer service reinvention, the company overhauled the way it looked at its customers, and adjusted product offerings to suit.
"It's crucial--whether it is an upsell or a cross-sell or an outbound effort--to know, does this information really impact sales?" says Janine Wilson, GBC corporate director of customer care. To improve profitability GBC has systematically examined its customers' needs more carefully to cut the fat from its product offerings. Hard, numerical analysis has played an important role, and Wilson says the company's call center provided the vast majority of the required data.
While field sales and third-party providers helped GBC create a more complete picture of customer needs, the call center was uniquely poised to challenge conventional assumptions about the business. "I think some companies forget to ask their customers, and one of the things GBC had done wrong historically was rely on 'this is what we think' internally," Wilson says. Executives had long assumed that clients with higher travel budgets would create more presentations and, therefore, had a greater budget for presentation consumables. But once asked directly, customers dispelled that notion, and GBC learned not to set sales targets based on travel.
John Georgesen, director of marketing sciences for business process provider Convergys, agrees: "Incorporate the voice of customer feedback, because it is so crucial for any true return on investment."
Convergys has worked with clients to analyze the impact of proposed changes in its approach to customer contact, both through predictive modeling and survey data. Georgesen says that often, the "gut feeling" driving organizations to a proposed course of action is based on mistaken assumptions. In one notable case Convergys advised a shipping company that the faster answer times it sought would not boost customer satisfaction as planned--averting a seven-figure investment in call center staffing. The key was talking to customers about what truly mattered to them.
"We built a lot of models, and helped them understand that what was driving satisfaction was the perception of the agent, and whether the call was being resolved [the] first time," Georgesen says. "Minimally was 'time [waiting] to speak to an agent' an important part of that experience."
Draw Lines in the Sandbox
The call center provides both excellent theory and practice for customer segmentation, as pressure is already high on service organizations to minimize costs, particularly for perceived low-value clientele. Customer segmentation is not a one-time event, and even performing initial categorization may yield some surprises or outlying results that companies can use for sales, marketing, and service efforts.
Be prepared to challenge your assumptions about your customer base. Segmentation should be a learning process. "Many customers fit in a nice clean bucket, but there are a lot of customers that are in between the lines, or migrating up or down into a new space," says Mark Davis, vice president of customer service at Pitney Bowes. "When we better enable the call center to segment customers in real time, it will provide better perspective into what the clients are doing."
Unify the Call Center and the Field
The call center should act as more than a center for dispatch and lead generation. Technology allows the call center to be a vital hub of data long after the field sales staff's route plan has been issued. Consider the approach employed by Pitney Bowes, which is currently replacing a legacy contact center hub system dating to the late 1980s with a modern, wireless, Siebel-based software system.
"At the fingertips of the 2,000 [call center] reps, we are going to have the exact environment the customer is in--not only the product types, but some of their infrastructure, the service history and replacement history of the products, [the status of] replacement parts, and billing history," Davis says. That will be possible only because field agents responsible for maintenance and repair will be able to update customer records from the field, acting as the eyes and ears of the organization during service calls.
Having more information about customer situations inside the call center means more first-call resolutions at the contact center and fewer truck rolls. "The longest solution to the problem is to put somebody in the car and send them to the customer site, especially when they don't have the information to solve the problem when they get there," Davis says.
There is little that goes on in the call center that doesn't benefit sales to understand, including information on returns or incremental purchases. At the start of the pipeline, the contact center can boost the value--and results--of lead-generation activities through the application of some basic metrics and judgment calls, by agents, passed along to the sales organization.
"As far as helping field sales, that's clear--show the field which leads are hot, so they can better focus resources," says Anne Milley, director of analytics strategy for SAS. "You need to be able to look at a set of communications, classify them, and follow to see how accurate the classification was."
Discard Old Metrics, Break Old Patterns
One of the most important responsibilities for any call center serious about meeting its service and productivity goals is to be unafraid to discard old ways of looking at data that is no longer in tune with the center's changing demands.
That strategy has taken hold in GBC's outbound customer contact center, where quality now trumps quantity. "Some of the traditional measures that call centers have used in the past, we're not going to use," GBC's Wilson says. "No longer do we say, 'Great, you made 60 phone calls.' We care about tracking whether all calls are quality calls. 'Did you get to a decision-maker?' and 'How many pieces of critical information were you able to get?' are more important than 'I picked up the phone and dialed.'"
Data Is More Than Numbers
Staggering amounts of information pour into an organization from customers in the form of text--be it handwritten requests or notes taken by a call center agent during an interaction. Traditionally business analysis software has focused on patterns in numbers, but recent advances have made it possible to reach accurate conclusions based on text data.
Hewlett-Packard recently adopted SAS Text Miner software to assist in textual analysis, and already the tool is providing valuable insights into customer preferences and patterns. "Customers that were really loyal were talking to the call center about different things than customers that weren't so loyal, or customers that did not buy as frequently or in as high of volume," says Randy Collica, a senior business analyst for HP.
One of the simplest but most rewarding applications of text mining at HP involved lead classification based on textual notes collected from an initial call center contact. HP used the text mining software to divide new leads into cold, medium, and hot rankings. Once the leads were passed to the sales staff, the majority of those leads performed as the analysis predicted. "With about an eighty percent success rate, I was able to do that, which isn't too bad," Collica says.
Cross-Sell for Value, Not Volume
Cross-sell initiatives demand a lot of your contact center resources, so be sure that the programs are creating returns that justify the investment. That means not only presenting relevant offers, but doing so only when the extra time spent on the phone doesn't disrupt expected service levels. "Seventy percent of the cost [of a call center] is in the labor. If you are only getting two percent ROI for your [cross-sell] campaigns, it's quite possible that depending on the value of the cross-sell, you're increasing revenues, but increasing costs at the same time," says Darci Moore, partner with Adroit Consulting.
Moore cites as an example one company that used a sophisticated scheduling system to ensure that cross-sell offers were only made when they would not likely effect operational performance. Agent screens would only prompt a cross-sell offer if the center was meeting its operational target of 85 percent call answering within 30 seconds. "If service levels dropped below the 85 percent target, it would stop screen-popping, because the number one priority was to meet service levels."
Monitor Customer Experience
Look to the call center to provide more than simple statistics that can speak volumes about efficiency, but say little about customer experience. Zane Taylor, senior director of operations for Internet media server vendor NetScaler, keeps a close eye on the performance of his company's products in the real world as reflected by data generated in the call center.
Taylor coordinates closely with NetScaler's level 1 support provider Vinciti, as well as his own customer support center, to resolve support issues with the company's high-value product line before they become customer relations problems. "We look at resolution ratios by time line, which provides me with the ability to see if there's a problem," says Taylor, who must work closely with the product development team to ensure that any major issues discovered at customer production sites are corrected quickly. "If there is a reproducible issue and the ticket is open for more than 10 days, then there's a [product] problem and we need to get a new [software] release qualified and get it to the customer in the field."
Mark Dayton, vice president of marketing for technology certification provider KeyLabs, emphasizes the importance of sharing customer experience information with the rest of the organization, particularly if the experience reveals critical flaws in the product: "Data needs to get back into those folks' hands--you need to get it to the documentation team, the support team, and the requirements team." While the call center always has to manage the impact of product experience, ultimately those interactions will set the stage for the customer's perception of the company's ability to create valuable products.
Contact Executive Editor Jason Compton at jcompton@destinationCRM.com