IBM has been called many things over the years. The company also affectionately known as “Big Blue” has thought of as supremely dependable, maverick in its approach to its alphaWorks labs, a safe purchasing bet (nobody ever got fired for buying IBM) and, to balance, it has also been thought of as slightly archaic and behind the times now and again.
Keen to show us that it has its finger on the fashionable pulse of contemporary matters, the company has this week released news of an IBM computer-based analysis of billions of social media posts that has been engineered to predict a change on the horizon in women’s shoe fashions, with heel heights – currently nosebleed-ingly high – apparently poised to decline.
The IBM project highlights the predictive capabilities of social media analysis as a source of valuable insight that can help drive business strategies and results.
According to a recent IBM study, chief marketing officers around the world recognize the need to embrace social media in their business – but only a small minority of them actually factor information from blogs and other forms of social media into their strategic decisions.
“Usually, in an economic downturn, heels go up and stay up – as consumers turn to more flamboyant fashions as a means of fantasy and escape,” said Dr. Trevor Davis, a consumer products expert with IBM global business services. “This time, something different is happening — perhaps a mood of long term austerity is evolving among consumers sparking a desire to reduce ostentation in everyday settings.”
A look back at the last 100 years of shoe fashion trends reveals that heel heights soared during the most prominent recessions in U.S. history. Low-heeled flapper shoes in the 1920s were replaced with high-heel pumps and platforms during the Great Depression. Platforms were again revived during the 1970s oil crisis, reversing the preference for low-heeled sandals in the late 1960s. Following that period, the low, thick heels of the 1990s “grunge” period gave way to “Sex and the City”-inspired stilettos following the dot-com bust at the turn of the century.
Here’s how the analysis was conducted: first, IBM used special analytics software to search billions of social media posts to identify individuals discussing shoes. This initial category contained tens of thousands of posts. Next, the software narrowed the list down to those who are key online influencers in the area of footwear – bloggers, for example. The software relied on algorithms that rated the popularity of these influencers by zeroing in on the ones who sit in the centre of large social networks – that is, writers of blogs that many other blogs link to and which in turn link to many blogs.
The IBM project illustrates how sophisticated analysis of social media could be used by manufacturers in planning future products, by retailers in choosing which products to stock and by marketers in planning advertising campaigns.
It could also help a city or government better serve its constituents.
The ability to analyse social conversation in real-time can help officials see how constituents are responding to policy decisions or how outreach could be varied across different channels to get the word out about specific events. Social media analysis could also serve as an early warning system for governments around special events and unexpected occurrences.
For example, public safety officials could use this technology as part of a rapid response system for flooding, earthquakes and other natural disasters; or to identify areas of public services delivery that need improvement.
Is this a pre-Christmas silly season story from IBM? Well yes and no, it’s a bit of lighthearted Friday tech trivia at worst. Is social media analysis becoming a) more prevalent and b) more seriously considered as a marketing tool overall — the answer must surely be yes. Even the Tweet that will support this story will be tracked by Apollo Research, as the company compiles its weekly compendium of technology writers’ news items.