Wise Guide
Financial success for an enterprise can boil down to what employees--the most valuable internal business resource--know. How to effectively, efficiently push relevant information to customer-facing employees is one of the goals of a knowledge management (KM) initiative. Properly leveraged, KM can, through in-house collaboration and participation, provide customers with prompt answers to queries without the assistance of a live rep, and can boost CSRs' ability to quickly track down answers for customers, strengthening service experience and customer satisfaction while lowering customer care costs. KM, as described by Jonathan Spira, CEO and chief analyst at research firm Basex, is "getting the right information to the right people, at the right time."
Organizations like Secure Computing, a provider of enterprise gateway security, are prime examples of companies whose KM deployments are reaping significant benefits. "We're saving the company--with our knowledge base and knowledge management program--between $250,000 and $350,000 a month," says Kris Solberg, manager of support tools at Secure Computing, a Talisma customer. "It's having relevant articles, accurate articles, a wide variety of content--and it's word of mouth. People are spreading the word that there's good self-help out there."
KM is a big, broad slice of an enterprise's processes. Here, we concentrate on industry pundits' responses to common KM mistakes that companies make in three areas: adding information into a knowledge base, putting guidance around content customers create, and deciding what information to keep internal and what to expose to customers.
Search Smart
TMI--too much information--has become a handy acronym for those moments when someone inappropriately blurts something out. According to some CRM industry observers, TMI is also applicable to KM practices. In a report published by Service Excellence Research Group (ServiceXRG) that includes findings of a study of 117 companies' KM processes, average clutter rate is identified as an indicator of how often content in a knowledge base is actively used. "Knowledge Management--Strategies, Benchmarks and Best Practices" also states that the respondents' average clutter rate is 56.6 percent, which means, according to the report, that more than half of all content housed in a knowledge base is inactive.
Tom Sweeny, a principal at ServiceXRG, attributes much of the existence of business data clutter to mandates and incentives for creating content. "The quantity of content is taking a bite out of the quality. Organizations that have made a commitment to create content for their knowledge base are going overboard, incenting and driving individuals to contribute and push that content into the knowledge base just as quickly as it can be identified and created. [But] it's all about understanding what information is in demand."
Another school of thought is that organizations worry too much about entering too much information into an enterprise's knowledge base; the theory is that the misstep is in leveraging shoddy search capabilities. "If you don't have a good search system there's no point in having a knowledge base to begin with," Spira says. "I would worry about making sure that anyone--whether it's a call center worker or any knowledge worker-- can find what they're looking for quickly and easily. That's the antidote to worrying about what should go in or not."
Part of the issue is assuming that users typing in a query can ask the right question of the technology to retrieve the needed information and failing to frame a knowledge base's input content in a way that customers understand. "Whether it's the capabilities of the search engine itself or the skills of the individual who is searching, there's a big gap between being able to find the right information and sorting through all the other information that's in there," Sweeny says.
Structuring content for relevancy and usability is the first major step in porting content to the knowledge base, according to Kate Leggett, director of product management at KANA. The second big step is to use best practice methods to help guide users to the right content, which goes beyond simple search techniques. "This includes practices like adding questions in the customer's vernacular and associating questions with one or more solutions, using case-based reasoning and expert reasoning techniques to surface answers to a user's question, and adding decision trees and clarifying questions to guide users to the right content."
However, Doug Warner, director of end-user experience and research for RightNow Labs, maintains that a huge flaw lies in assuming that users will locate desired content by searching. Instead, he says, the preferred choice is browsing. (Searching, according to Warner, refers to user behavior for finding information when entering textual queries into a search engine, while browsing refers to user behavior for finding information in the absence of textual queries.) RightNow's analysis of several different sites and more than 700,000 different user sessions revealed that just 20 percent entered search queries, while 80 percent turned to browsing to get content they were looking for, Warner says. "Browsing is supported via such things as top-viewed documents, recently updated documents, related documents, and so on. Browsing is much easier than searching, especially when one is not an expert in a domain."
Public Access
A growing number of businesses have accessed more collaboration-driven initiatives, such as customer forums. Tim Hines, vice president of product management of Consona CRM, contends that customer-created content should only have minimal and basic guidelines assigned to it. "By openly collaborating with the customer in a Web 2.0 environment--rather than putting strict guidelines on their feedback--enterprises can have a first-hand view into their internal CRM processes. Customers can offer invaluable insight into those products and/or services that are running effectively or not so effectively."
Perhaps the biggest challenge customer communities face is helping customers determine the source and credibility of posted content. One method that Web sites like Amazon.com and Target.com are leveraging to provide site visitors with a better sense of the source and reliability of the content is to incorporate author rating systems. Customers can write an online review that is seen by other customers, and customers have the ability to rate whether or not a review was helpful. This dimension essentially helps the community begin to regulate itself, according to Sweeny. Then, for example, if a large number of visitors tack a review as not helpful, a company can take steps to help reinforce the source, authority, and the credibility of that content.
"You can leverage the customer community and resolve issues through collaboration or other forms of contributing to the community and the knowledge base in general," says Mark Buckallew, InQuira's senior director of products. It can be valuable to customers to see what their peers are doing to resolve problems or what others thought about particular products, especially while readers themselves are making purchasing decisions. "That's very useful information to be able to display, and helps to drive usage of the site."
Secure Computing lets customers create content by providing forums for select products. These forums are searched periodically by the business to see what the hot topics are, and then customers are directed to the company's knowledge base for company-sanctioned answers, according to Solberg. It also provides article-feedback functionality, enabling customers to reply whether or not an article solved their problems, and allows Secure Computing to tweak articles accordingly. "Customers can say things like 'Yes, this solved my problem,' or 'Step three needs a space in between this command and that command,' so we can fix things that way," Solberg says. "We also put in a feedback loop that says 'Here's an article topic I'd like to see covered.'"
Another component of the company's strategy is to run weekly the Unanswered Questions feature within Talisma Knowledgebase, which tracks and logs all unanswered questions and keywords for which customers search. If customer searches turn up no articles, the company can go back to its tech-support engineers to modify articles or create new entries.
While collaboration can help companies glean valuable customer insight, it's imperative not to let efforts around fostering open communication override initiatives to ensure content is accurate, whether that information is posted on a forum or actually added to a knowledge base. By not establishing policies around what information users can and cannot post, companies are not setting their knowledge base up for success, Leggett says. "Companies make the mistake of not routing the content to be added to a knowledge base through a configurable workflow engine so that content can be reviewed for accuracy, relevancy, uniqueness, and findability prior to being added to the knowledge base."
Do You See What I See?
As companies concentrate on beefing up information lodged inside their knowledge bases--with content generated by internal experts and customers--they must also determine what information should and should not be made publicly available, such as proprietary product information. However, part of the issue surrounding how far into the internal processes customers should be allowed hinges on corporate culture, Sweeny says. "There are companies that are very open and they make it very clear with their customers that you know everything we know and we want to hear from you to make it better," Sweeny says. "The other end of the spectrum is to keep a very controlled, very regulated message." Part of it, he says, comes down to "Why have you done this? Are you trying to offload costs? Give the customers everything you know? Or are you trying to basically use the content to drive a very specific end result?"
Buckallew says one way to go is to move content used to solve customer issues to the Web site as quickly as possible. "The theory is that if they come in to the call center these are new problems that exist and there are not existing solutions out on the Web site. So the faster a company can get things out on the Web site, the more effective the self-service is." A word of caution: Publish content, but do not overlook due diligence to ensure accuracy and relevance.
It's important to note that as some companies continue to develop their KM initiatives, tailoring the service experience is rising in importance. Sweeny puts it this way: "Until we start changing the user experience to reflect the differences of the diverse population we're serving, we're really not going to achieve a level of effectiveness beyond the shotgun blast of here it is, help yourself, good luck finding what we published."
Contact Associate Editor Coreen Bailor at cbailor@destinationCRM.com.
Meta Facts
FAQs can provide customers with answers without help from a live agent, But if those questions and answers are simply thrown on to a Web page, they'll miss the mark. Have a look below for three frequently asked questions about--yep, you guessed it--faqs.
What kinds of questions are best suited for FAQ pages?
As the letters in the abbreviation suggest, FAQs are ripe for answering the most common questions from an organization's customer base. Responses to posted questions should be easy to understand and crisp, yet fully composed. Consider links to supplementary pages that feature more comprehensive information. It is important, however, not to view FAQs as a one-shot deal. Look at "what are they asking today versus what did they ask [you] yesterday and adjust your FAQ list to really meet those needs of what they're looking for today," says Doug Warner, director of end-user experience and research for RightNow Labs.
What kinds of questions are not appropriate for FAQ pages?
Questions with complex solutions that require multiple steps that may not apply to all users, and questions and answers needing interaction with a support agent to recommend the best path to troubleshoot a device, are examples of queries ill fitted for FAQ pages, according to Kate Leggett, director of product management at KANA. Questions including any proprietary or competitive information are also red flags.
What are some of the biggest mistakes companies make when crafting FAQ pages?
Assuming what a company's oft-asked questions are rather than assembling FAQs based on actual questions from customers is a surefire way to a low-traffic FAQ page. Failing to group FAQs by product or problem category and not using dynamic learning to update FAQ lists are also major gaffes, according to Leggett. --C.B.
Information Overload: A Knowledge Nuisance
When was the last time you made it through a task on your calendar without getting interrupted by phone calls, emails, the Internet, or message pop ups? Thanks to unnecessary workflow interruptions and time needed to refocus, knowledge workers are losing an average of 2.1 hours per day, costing U.S. businesses $588 billion annually, according to research firm Basex's study "The Cost of Not Paying Attention: How Interruptions Impact Knowledge Worker Productivity."
Jonathan Spira, CEO and chief analyst at Basex, defines information overload as the provision of information in excess of an individual's cognitive or emotional capabilities to process and comprehend. "Information overload can impact contact center agents in the course of their work, making it difficult or impossible to find the information that they need to complete a call or answer an email, possibly causing them to provide incorrect or incomplete information to a customer."
Spira contends that knowledge workers must get a better handle on what interruptions require immediate attention. "Knowledge workers need to learn to recognize the trap of the tyranny of the convenient," the report states. "And they need to understand what's important versus what's urgent and place their work in perspective with what others might be doing." --C.B.