Comments:"IBM's Watson Now A Customer Service Agent, Coming To Smartphones Soon - Forbes"
IBM’s question-answering Watson supercomputer is building quite the résumé. First it won a much-publicized showdown against the two greatest Jeopardy! champions of all time, then it went to medical school and emerged as a budding oncologist. Now Watson has a new job–as a customer-service agent with the mostest. The help desk is a bit of a step down from fighting cancer, but IBM is nothing if not pragmatic. U.S. organizations spend $112 billion on call center labor and software, yet half of the 270 billion customer-service calls go unresolved each year, presenting a fairly sizable opening for an enhanced cognitive computer. Let’s face it: Rare is the occasion when you a) reach a live person and b) they know what they’re talking about. Why not give silicon a chance?
Starting in the next few months, IBM will be rolling out with several key customers an “Ask Watson” feature that will greet and offer help through various channels: Web chats, email, smartphone apps and SMS. Some customers will eventually equip the service with voice recognition from a partner such as Siri or Nuance. The guinea pigs include Australia’s ANZ Bank, Nielsen, Celcom, IHS, and Royal Bank of Canada.
Each of the beta customers will come up with their own interface and branding for the service, and many will begin with internal employee use only, but for some this will be the first time the general public will get to query Watson directly with questions such as “Hey Watson, how do I diversify my 401(k)?” or “Hey Watson, what are the flights between Dallas and the West Coast that I can book with miles?” or “Hey Watson, how many gigabytes do I have left on my mobile plan?” IBM says we’ll start to see the first Watson-powered consumer apps in the second half of the year.
According to IBM, close to two-thirds of the 135 billion unresolved calls each year could have been resolved with better access to information, the search for which consumes six to eight minutes per call, on average. At each of the companies testing out Watson, the computer will be fed loads of product information from closely held databases such as catalogs, training manuals, product disclosures, terms and conditions, emails, customer forums, and call center logs, as well as publicly available feeds and reviews from places like Amazon, Yelp, Trip Advisor and technical support communities. In IBM’s tests using its own call centers and proprietary data, Watson delivered a 40% reduction in search time for information. “Watson pulls up stuff that an agent wouldn’t because it is looking for semantic links, not just doing text-matching based on keywords,” says Manoj Saxena, general manager of IBM Watson Solutions.
ANZ Bank is going to start deploying Watson at its private wealth group, beginning with insurance offerings. “Imagine if you could sit down with an adviser and, in the time it takes to make a cappuccino, Watson will pull up all of your accounts, read all the fine print, and tell you what kinds of insurance protection you’re missing or where you’re overcovered,” says Joyce Phillips, CEO of ANZ’s global wealth and private banking group. ANZ first discussed the idea with IBM a year ago. Phillips, frustrated by the lack of innovation in the private wealth business, is most looking forward to using Watson to capture a new customers’ financial picture in an informal Q&A session to avoid the hassles of making them fill out a personal financial statement. If a customer signs up with the bank, Watson can theoretically monitor all the changes to the bank’s products and assimilate and correlate them with a client’s financial picture. Skeptics will scream this is a death knell for the human investment adviser, but a good adviser would be happy to get that kind of back-office help.
Nielsen, famous for its TV ratings, intends to embed Watson in its software and service used by Madison Avenue media planners who buy TV, radio, print and Internet spots based on audience metrics such as reach, cost and frequency. They’ll be able to query the system with basic or complicated questions about how much to spend on which medium to meet their campaign goals. “Most of the media planning tools out there can only mine a specific data set,” says Randall Beard, global head of ad solutions at Nielsen. “But Watson will know about a research paper written by the Ehrenberg-Bass Institute for Marketing Science in Australia that no 24-year-old media planner in New York City might have seen.”
Teaching Watson to play championship-level Jeopardy! took IBM four years. Teaching it to deliver reasonably accurate answers to diagnostic questions took only four months. Teaching it to be a 24-year-old media planner shouldn’t take nearly that long.