[Paper Review] Style in the Age of Instagram: Predicting Success within the Fashion Industry using Social Media

Jaehyuk Park, Giovanni Luca Ciampaglia+1 more|KDI Central Archives (Korea Development Institute)|Aug 18, 2015

Digital Marketing and Social Media28 references21 citations

TL;DR

This study develops a machine learning framework to predict early career success in the fashion industry using physical, professional, and Instagram social media metrics. It finds that social media presence—particularly active engagement on Instagram—is a stronger predictor of model tenure than traditional industry factors like top agency representation or physical aesthetics.

ABSTRACT

Fashion is a multi-billion dollar industry with social and economic implications worldwide. To gain popularity, brands want to be represented by the top popular models. As new faces are selected using stringent (and often criticized) aesthetic criteria, \emph{a priori} predictions are made difficult by information cascades and other fundamental trend-setting mechanisms. However, the increasing usage of social media within and without the industry may be affecting this traditional system. We therefore seek to understand the ingredients of success of fashion models in the age of Instagram. Combining data from a comprehensive online fashion database and the popular mobile image-sharing platform, we apply a machine learning framework to predict the tenure of a cohort of new faces for the 2015 Spring\,/\,Summer season throughout the subsequent 2015-16 Fall\,/\,Winter season. Our framework successfully predicts most of the new popular models who appeared in 2015. In particular, we find that a strong social media presence may be more important than being under contract with a top agency, or than the aesthetic standards sought after by the industry.

Motivation & Objective

• To investigate whether measurable physical and professional characteristics of new fashion models can predict their success in upcoming seasons.
• To assess whether social media activity—specifically Instagram engagement—improves the predictability of model success beyond traditional industry metrics.
• To understand the role of social media in disrupting traditional, aesthetics-driven casting processes in the fashion industry.
• To examine whether social media presence acts as a proxy for offline industry buzz and information cascades.
• To contribute to the broader 'Science of Success' by analyzing how collective attention on social media influences cultural market outcomes in fashion.

Proposed method

• Collected data from the Fashion Model Directory (FMD) for new models entering the 2015 Spring/Summer season.
• Gathered Instagram activity metrics (e.g., follower count, post frequency, engagement rate) for the same cohort of models.
• Constructed a machine learning model using supervised classification to predict model tenure (number of runways walked) through the 2015–16 Fall/Winter season.
• Employed regression analysis to assess the association between tenure and individual features such as height, body type, agency reputation, and social media activity.
• Used evaluation metrics such as AUC-ROC and F1-score to assess model performance on held-out test data.
• Controlled for cumulative advantage effects by focusing exclusively on new entrants, minimizing bias from pre-existing fame.

Experimental results

Research questions

• RQ1 Given data about measurable physical and professional characteristics of a model, can we predict whether she will be casted for the upcoming fashion season?
• RQ2 Does the addition of relevant signals of social media activity improve the predictability of success of models?
• RQ3 To what extent does social media presence act as a proxy for industry buzz and information cascades in model selection?
• RQ4 How do traditional aesthetic criteria (e.g., height, thinness) compare to social media metrics in predicting model tenure?
• RQ5 Can social media activity serve as a reliable early indicator of career success in the fashion industry?

Key findings

• Social media presence on Instagram was a stronger predictor of model tenure than traditional factors such as top agency representation or physical attributes.
• Models with active Instagram accounts were more likely to walk in multiple runways, even after controlling for other variables.
• The machine learning model successfully predicted the majority of new models who achieved success in the 2015–16 season.
• A strong correlation was found between social media buzz and offline industry visibility, suggesting that online attention amplifies real-world opportunities.
• Even among new models, the winner-takes-all dynamic was evident: 24% of models walked in at least one top-tier runway, and this single event dramatically increased their career visibility.
• While physical traits like thinness and height were significantly associated with higher runways, their predictive power was outperformed by social media metrics in the final model.