Using Social Media to Predict Fashion Trends (Guessing is so 2009!)

Social media monitoring is generally dominated by Marketing, PR and Communications, so it was refreshing to meet Geoff Watts from Stylesignal earlier this month. Geoff and his team...

Stylesignal_worldpulseSocial media monitoring is generally dominated by Marketing, PR and Communications, so it was refreshing to meet Geoff Watts from Stylesignal earlier this month. Geoff and his team have developed a social media monitoring tool for the specific purpose of tracking and predicting new fashion trends – also known as “fashion forecasting”. By tracking the websites, blogs, Tweets and images published by a selection of influential fashionistas, Trend Science is able to provide uniquely valuable insights into a notoriously unpredictable market, spotting which colours, cuts and shapes are going to hit the high-streets before the herd has caught on. It’s pure genius.

The functionality itself is rudimentary by comparison with some of the more serious (and costly) monitoring and measurement tools but, in recompense, this makes it extremely user-friendly. You can check out what’s trending by looking at World Pulse – a list of fashion terms, including labels, ranked by current popularity. Or you can use Zeitgeist to run bespoke searches for specific keywords, which produces a list of the latest posts, tweets, images etc. for that term. Alternatively, you can pick one of the pre-configured categories to plug into the latest “neckline” or “make-up” chatter, for example. There are a LOT of people talking about these, though, if I’m honest, neither floats my boat particularly.

While Trend Science does produce attractive lists, charts and gauges, Geoff was at pains to explain to me that Stylesignal’s real skill is in predicting trends through offline analysis.  They really use Trend Science to validate their industry predictions and back them up with quantitative data. I think this is a healthy approach to social media monitoring. It’s best not to see it as a solution in itself, more a useful adjunct to existing research and analysis tools.

The most fascinating thing about Trend Science is probably that it’s genuinely forward-looking. David Cushman of the 90:10 Group made the interesting point at Monitoring Social Media 09 (and on his blog) that monitoring past data to predict the future is somewhat oxymoronic. Yet Stylesignal is doing just that, and to good effect. Perhaps the fashion industry, being highly influencer-led, is uniquely positioned for this kind of real-time analysis, but I suspect there are other industries that could be similarly pre-disposed: music, for one.

Lastly, it’s worth mentioning that Stylesignal’s solution is yet another example of social media monitoring tools going niche. Several weeks back I wrote about a niche monitoring service for the medical industry, and I can only see this trend continuing. Somewhat sneakily, I tested Trend Science by searching for “striped pyjamas”. Because the data sources are highly targeted the results eschewed any mention of books, movies or the holocaust and produced a short list of tweets and blog posts about pyjamas. Full marks again to Geoff and his crew.

Read Next

In this article

Join the Conversation

7 comments

  1. Macala Wright Reply

    Interesting post. I think real-time fashion trends are helpful to retailers on a local/regional level. I also think brands that cater to fast fashion trends would find the data valuable. Trending in itself is completely different. Most brands and retailers are already planning through 2012. Services like StyleSight and Accessories Directions are started. Those thing aren’t going away any time soon. Our industry is still very traditional despite large strides forward in the past two years via social media. Influencers will still lead trends way out as well as what’s taking shape in real time – sites like Lookbook.Nu, Chictopia and Weardrobe are daily inspiration for current trends and very visual.

  2. Luke Brynley-Jones Reply

    Thanks for putting this into perspective Macala. Localisation is something I hadn’t really picked up on, but I can see that would be valuable.

  3. Macala Wright Reply

    You’re welcome! This is actually really neat. I can’t wait to learn more.

    @Macala

  4. MO123 Reply

    Hey
    Is there any way i can zoom into the graph?
    Thanks

  5. Making a Statement | Fashion Dimensions Reply

    […] media is responsible for much of what the public sees and hears regarding popular trends. This can also […]

  6. Anonymous Reply

    Really good article you have shown above but time and fashion both going so far. We can’t predicting about fashion because its changing every new day.
    http://www.attitudeinc.co.uk/
    street wear

0 Shares
Share
Tweet
+1
Share