BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Companies use data analytics to understand customer trends, but they also cut through those trends with human discretion to make business decisions. JackThreads, for example, uses customer feedback data to adjust customer service -- or sometimes to proceed with a sale even if the data might suggest canceling it.
JackThreads sells discounted, private-label menswear online. That digital-only storefront gives the Columbus, Ohio, business a better understanding of personalized customer feedback from social media, online chat, emails and phone calls.
Tracy Biddle, JackThreads' customer service operations coordinator, said the company last year streamlined the gathering of customer comments by using the same terms -- satisfied, angry, among others -- across the board so employees would have a consistent baseline of customer sentiment.
The effort is just one of many steps JackThreads has taken to harness direct, and sometimes brutally honest, customer feedback. For instance, after only "dabbling" with the company's online Web chat offering, JackThreads recognized that it could not just help correct customer service problems, but also spur sales, Biddle said.
"We had used our chat service to look up transcripts [of customer calls] and get an idea of what sort of words customers were using in conjunction with others," she said. For example, Biddle and her team wanted to get to the bottom of the company’s high rate of product returns. "We dug into customer interactions, and we learned that 60% of customers were contacting us because sizing wasn't as expected. So we were able to go back and re-evaluate our sizing, and customers were able to pick out the right size the first time without having to exchange."
Biddle culls customer experience data from the customer service and support ticket software Zendesk Inc., chat transcripts from LivePerson and social sentiment from Conversocial. She merges all that information in Excel.
Sales data analysis offers a view into customer purchases and returns, but Biddle finds that a review of sentiment from chats and social media can point to trends the raw data may not reveal. "We might have excitement around a product that's only $20. Sales [data] won't show a huge fluctuation based on that price point, but if it's getting website clicks or social media attention and it's not a huge profit item but something the customer appreciates, then it's something we want to continue to offer."
JackThreads also sees value in having customers' unvarnished voices guide the data mining. "We’re able to use the data to better the customer experience, but by caring about the voice of the customer, we're also able to dig back into the data and figure out where we missed something," Biddle said.
"We have an internal blog that we use to [share] customer sentiment. So instead of me rewording [sentiment] for the general manager, we provide customers’ wording. We copy and paste exactly what they said. There is something to be said about stepping away from the traditional data and looking at what customers are saying."
Education analytics helps college, students succeed
There's still a place for social media analytics in business