Melbourne Spring Fashion Week has been a big success in social media, particularly in 2015, and a showcase for IBM and its IBM Studio & Interactive Experience talents and platforms. Surprisingly, the current 2016 event has fallen short of social media expectations. Our research shows that a different role for images (and Instagram) may have played a part in the diminished social media activity this year.
Activity well down in 2015
Is there a problem? The numbers don’t lie. The 2015 MSFW ran from 28th August to the 4th of September, in 2016 from 26th August to 2nd September. As of today the 2016 festival has 2 days to go.
For the equivalent period in 2015 (2 days to go until the finish) there were 4,392 tweets (#msfw). In 2016, 1857 tweets. That is, 2016 is running at 43% of the 2015 Twitter volume.
Visually this is clear comparing the activity chart for the two events.
The same can be seen in the Twitter Reach statistics.
The reach for 2016 so far is 10.0m estimated impressions (left), and in 2015 for the whole period it was 38.3m estimated impressions (right). Even though the 2016 festival still has a day and a half to go it is unlikely to reach anywhere near the reach of 2015.
The story might be found by semantic analysis, by comparing the buzzgraphs for each period as the buzzgraphs show the words which are most often used with each other and give insights into what is capturing attention.
By comparing 2015 with 2016 there are some noticeable differences, in particular that in 2016 words associated with images are missing e.g. “instadaily”, “picoftheday”, “periscope”. These concepts of images and live broadcasting are missing in 2016, although IBM “Watson” gets a mention – probably because the volume of other activity has dropped off and IBM has been actively promoting its participation in the event with the #CognitiveCouture theme.
The most popular tweets reflect the buzzgraph differences as well. The 2015 most popular tweets, at left, are about the fashion.
The 2016 most popular tweets , below, are about IBM.
What other clues are there that images and Instagram has played a smaller role this year? Instagram does not allow social media monitoring tools to search retrospectively by date, so it is not possible to compare the activity between 2015 and 2016. But we can still explore some date-specific hashtags.
Using the hashtags #msfw15 and #msfw16 there were 1826 images in 2015 versus 1042 in 2016 (56%).
For the hashtags #msfw2015 and #msfw2016 there were 937 in 2015 versus 599 in 2016 (64%).
Those two ratios are higher than the ratio of Twitter activity in 2016 compared to 2015 but still significantly down on 2015.
Does the lower Twitter activity cause the lower Instagram activity, or vice versa? That we don’t know. But we do know that people love to share images, especially about fashion.
So while we may not be able to pinpoint a definitive cause there is reason to suggest that in the post-mortem of the MSFW’s performance in social media that the role of images and Instagram be examined.