In its second-largest acquisition, Amazon buys the company for $970 million.
Natural language processing could save large retailers $400,000 per year, but the technology is not yet widely used.
Only one-quarter of companies recently surveyed by Jupiter Media Metrix use Natural Language Processing, an intuitive language program that identifies keywords to resolve customer inquiries without service representative intervention. But incorporating the technology to automate e-mail responses where possible – such as on inquiries about order status -- could save as much as $90,000 a year for companies that receive 20,000 e-mail inquiries per month, and up to $400,000 per year for those that get 70,000 messages per month, says David Daniels, Jupiter analyst.
Jupiter found that the retail industry is providing the timeliest e-mail customer service, but even so, only 53% of retailers are meeting customers’ expectations of a response within six hours. Across several industries, only 38% of companies responded within six hours, while 33% take three days or longer to respond or don’t respond at all. The second timeliest e-mail response came from the financial services industry, where 46% responded within six hours, while the slowest responses came from travel companies, where only 12% made a response within six hours; and corporate brand web sites, where none of the companies surveyed made the six-hour deadline.
Jupiter also found that e-mail auto-responses that don’t use natural language processing technology were rarely effective – only 2% of them successfully resolved customer inquiries. “Over the last several years, companies have bought into vendor claims that e-mail automation can lower costs and raise efficiencies,” says Daniels. “However, recent Jupiter research indicates that e-mail automation has not been perfected and will not be until NLP is integrated.” The incorporation of the technology, he estimates, could reduce customer service representative staffing levels by as much as 40% for companies that handle large volumes of e-mail.