In its second-largest acquisition, Amazon buys the company for $970 million.
(Page 2 of 3)
While functions such as product research, order entry, order history and order status were some of the first customer service functions to be automated successfully online, now new technologies are helping to push more complex tasks into the realm of self-service. These self-service options are most often linked to a knowledge base, a behind-the-scenes repository that centralizes product and customer transaction data. The knowledge base, the backbone of the soup-to-nuts offerings of major CRM vendors, serves up responses to customer self-service queries; it’s also accessible to call center agents seeking information on customers’ behalf.
Natural language search technology is letting both self-service customers and call center agents make better use of knowledge bases. It’s formatted to deliver answers in response to queries posed as questions rather than keywords. Among other functions, according to Creese, the software adjusts the search function to allow for typos, misspellings and variances in query language. It can pull relevant data from both documents and databases to find answers for site visitors.
While it allows for differences in how users pose questions, upping the chances they’ll find the information they’re looking for, natural language search still delivers multiple answers. When resolving a customer’s question requires one answer instead of a multiple choice, guided assistance is another approach.
At CRM technology provider eGain Communications Corp., guided assistance is part of a suite of CRM solutions. “Search gives you a lot of options to choose from as your answer. Guided help takes you through a series of questions and shows you the right answer,” says Anand Subramaniam, vice president of marketing. Online guided assistance is a Q-and-A session, in which the online help function displays a question at the user interface and then posts additional, preloaded questions triggered by how the customer responds. Each question-and-answer round draws a tighter circle around the universe of possible answers until the customer is left with a final answer or recommendation.
While automated, guided assistance can help online customers with tasks that range from picking a camera to configuring a computer, self-service has its limitations. At some points in the self-service process, customers are better served with the opportunity to escalate to live, human assistance, whether by live chat, e-mail or phone.
Dealing with limitations
One of those points is when customers are confused about the self-service process itself. The live agent’s job on escalation is to solve the customer’s problem; it’s also to try to guide them back to self service, where appropriate. Athletic shoe manufacturer Fila Holdings S.p.A, an ATG client, has been able to reduce its call center cost by almost 20% by deflecting phone orders from small retailers to online order entry at its B2B web site. Fila also requires call center agents to supply relevant URLs to customers ordering by phone to let them know where they can get the same information online.
When call center agents using Kana’s Kana IQ knowledgebase respond to customer questions by e-mail, they include links to those answers in the knowledge base. “This says to the customer that next time, they might be able to find the answer themselves. The agents don’t actually say that, but when you provide the link, customers can easily pick up on it,” says Jochims. Co-browsing, in which the agent takes control of the user’s browser to help the user navigate the site is so far little-used in e-commerce, but it can do the same thing, essentially; training customers to serve themselves.
Because customer interactions do cost more at the contact center than online, site operators are developing rationing strategies that escalate customer service from self-help to human help in the situations where it will have the greatest impact. Executives at CRM solutions provider Talisma Corp. describe three types of escalation strategies: by offering live help for visitors to certain predefined pages on the site, according to visitors’ pathway through the site, and according to the trigger of defined events.
For example, “You can offer an expensive service channel such as chat on a set of pages which are of high value, maybe a page that sells cameras or camcorders,” says product manager Amit Bansal. The second strategy is based on customer behavior. “Based on how the customer navigates the site and the URLs the customer has clicked on previously, for example, we can show a catalog page. The third way is based on events. For instance, if the customer clicks on a cart and then abandons it, it could trigger a chat session.”
Like a store shopper
Tim McMullen, Talisma’s CMO, likens the company’s multi-tiered CRM approach to the experience of a customer walking into the corner store. “When you’re at the store, they are monitoring your behavior before they walk up to you and ask if you need help,” he says. “That’s human intervention into self service in the retail experience.”
Yet another way to decide when to move customers beyond online self service is to set the parameters based on a customer’s value to the company or the value of the transaction. “If I am sitting on a shopping cart and have an immediate question I need to answer, chances are I won’t keep the shopping cart open as I go through a standard process of submitting an e-mail question or making a phone call, and I’m going to abandon my cart,” says Kana’s Jochims. To minimize such bailouts, particularly among high-value customers, he suggests, the site operator could recognize those customers on log-in to allow them instant access to an agent who will respond to questions immediately via live chat.
Quality vs. cost
While the dividing line between when to leave online customers to serve themselves and when to offer live help will differ for each online marketer, one constant principle offers some guidance. “The need for escalation is based on the limitations of your self service solution,” says Subramaniam. In eGain’s view, the success of online self-service as well as its future depends to a large extent on how it’s presented to users. Unless the service or information the customers seeks is offered in a way that’s useful, the customer is unlikely to try self service again.