The big data-driven, cognitive computer that has outdiagnosed some physicians and beat a champion on the Jeopardy game show is now being put to a different use: predicting what the big holiday sales trends will be and helping shoppers buy the season's "hot" gifts before they sell out.
IBM announced today that it has launched the Watson Trend App -- currently available just for iPhone and iPad users via the Apple App Store -- to help people better figure out what the biggest holiday shopping trends are, and how those trends might change over time.
The Watson app tracks online conversations across more than 10,000 sources, including social media, blogs, and product ratings and reviews, then applies predictive analytics to forecast whether a trend is a fleeting one or one that's likely to last through the holidays. The tool currently provides trend information for three major types of products: consumer electronics, toys and health/fitness products.
'Cognitive' Steps in Where Web Analytics 'Fall Short'
While the Watson app is aimed at shoppers, IBM clearly envisions doing much more with its famous cognitive computer as the holiday shopping season gets ready to kick off next week on the day after Thanksgiving.
For instance, the company has scheduled a Twitter conversation for Wednesday afternoon on the topic, "How cognitive will power the holiday season for both retailers and consumers." IBM Commerce is also sponsoring a December 2 webinar with Internet Retailer magazine on "Turning Commerce Data into Online Merchandising Insight."
"Today, online merchandisers manually compile information about products and categories while lacking the analytics and insight needed to effectively manage the customer experience and drive business results," according to the webinar's registration page. "Web analytics can help you begin to understand the customer but falls short of providing actionable insights needed to maximize business performance."
Cognitive computing can help businesses "optimize every opportunity for increased sales" with the help of visualization and listening tools that identify market trends and patterns, according to the information on the registration page.
Expect More Mobile Browsing, Buying this Year
IBM said its Watson Trend app provides a more accurate picture of what items are likely to sell best because its insights are based on evolving online conversations rather than on static "what's hot this season" lists compiled by various sources.
The app works by aggregating findings about conversations into three trend groups: content, context and sentiment. Each group is then given a relative daily Trend Score between 0 and 100 that is based on how many conversations are taking place as well as how quickly those conversations are growing.
Another prediction for this year's holiday shopping season is that more people will be browsing and buying via mobile devices rather than from desktop computers. Over the five-day shopping period from Thanksgiving to Cyber Monday, mobile traffic is expected to grow by nearly 57 percent while mobile sales are predicted to be 34 percent higher than last year.
So what is Watson currently predicting as some of the hottest items on people's holiday shopping lists this year? Leading the tech category as of this morning were the Apple Watch (trend score of 99) and Samsung TVs (trend score of 50), while Nike running shoes (trend score of 19) dominated in the health/fitness category and Lego Friends (trend score of 20) and Minecraft Legos (trend score of 16) held the top two spots in toys.
"At one point or another, most of us have used our social networks to help make a purchase decision," IBM said on the Watson Trend site. "Sometimes you get great advice, and sometimes not . . . With this app we are trying out the idea that Waston's insights can go beyond the usual gift guide to help users find and give more inspired gifts."
In September, IBM said that it would be building a new Watson Hub in San Francisco to bring its cognitive computing platform closer to startups, developers and venture capital firms. The company last year invested $1 billion "to accelerate the commercialization of IBM Watson, bringing cognitive computing to more clients and partners in more industries."