Fashion Brands and Social Media: Counterintuitive Evidence

Sankt-Petersburg, Russia, September 24, 2017: iphone 6s with icons of social media on screen. Smartphone life style smartphone. Starting social media app.

By Matteo Altobelli

Marketing budget has been exponentially going digital. In this article, the author explores the social media platforms being used by 108 fashion houses, the content they post by type, theme, format and how effective they are in engaging their followers.



While the luxury goods market has grown over the last year at a low single percent digit rate, the online part of it has grown by 24 percent.1 As a consequence, over the last few years, marketing budgets have been increasingly shifting online.

For a luxury brand, pursuing marketing campaigns through owning messages, images and media is definitely a preferred approach than buying digital ads. Brand integrity can be more easily preserved while avoiding the more opaque digital ad auction market. In this respect, a solid strategy on social media – which ones to use, what content, what frequency – has become crucial.

We analysed 108 fashion houses ranging from the biggest ones with revenues in the billion-euro range to niche ones with a few million-euro revenues producing handcrafted items. We focussed only on clothes, shoes and bags. The vast majority of the companies (73%) cover both women’s and men’s items, while 10 percent are dedicated to men’s items only and 18 percent to women’s goods only.


A Photo is Worth 1000 Words…

The one and only app used by 100 percent of the brands is Instagram. Instagram is particularly suited to snap and share image cum caption messages in a very image-driven industry such as fashion.

The analysed companies spread their digital marketing budgets and efforts over 19 different social media platforms from the most common to the very niche ones. On average, each brand is active on five social media platforms. The one and only app used by 100 percent of the brands is Instagram. This should not come as a surprise. The photo (and video) based social media is particularly suited to snap and share image cum caption messages in a very image-driven industry such as fashion. Likes and comments then allow for sharing and interacting with the brand.

[ms-protect-content id=”9932″]

The other three most used social networks are Facebook (96%), Twitter (84%) and YouTube (71%). Each of them specialise in a specific function, though not necessarily in an exclusive way. This four-app suite covers pretty much the most relevant needs to interact with customers. Instagram appeals to the need to physically view the different items while Facebook (on which, more often than not, pictures from Instagram are also repurposed) provides a more long text-based type of interaction. On top of that, Twitter provides the media for short announcements (e.g. events, deadlines) while YouTube provides a video-based medium with no duration limits for videos on in-depth coverage of events, corporate stories and the like.

The fifth most common social network is Pinterest used by 60 percent of the houses.  

Considering that China has been the most important source of growth for the luxury goods market in the last years and that major international social networks are normally blocked in China, it is rather surprising to see that only 17 out of the total 108 brands actually use Weibo, often referred to as the Chinese Twitter. Also, only eight out of these 17 firms use WeChat, the other Chinese app that provides many functions in one app (e.g. messaging, online purchasing, mobile payments, social network, online games) and that is heading toward one billion users.

Based on the above results, the study focussed on Instagram as the most relevant social media to analyse each brand on. So, an analysis of a few tens of Instagram posts for each of the 108 fashion houses followed. The objective was to review the type of content and format they use the most. Further to that, the various posts were split according to whether they were photos or videos, related to women’s or men’s products (or neither one) and finally according to the type of subject. 

Further on, engagement yielded by each post was analysed. As for the relevant metrics, the number of likes received by each post compared with the average of likes across all of the samples for any given fashion house was used. 


…but a Video is not Worth 1000 Photos

Within the time window of the study, 75 percent of the companies posted both videos and photos. The remaining ones used only photos. Out of the total posts, 88 percent were photo-based while 12 percent were in video format. When comparing engagement (number of likes per post/number of followers) provided by photos vs. videos, the former lead by 32 percentage points. On average, engagement on video-based content is below the average when compared with photos in 90 percent of the cases.

The pictures always deliver better engagement than the video. From a marketing budget standpoint, this should be taken into consideration as videos can be significantly more expensive than photos in terms of planning, casting, production and post-production.

In several cases, frames from a video were posted as pictures adjacent to the video; the pictures always deliver better engagement than the video. From a marketing budget standpoint, this should be taken into consideration as videos can be significantly more expensive than photos in terms of planning, casting, production and post-production. 

At first sight, this might seem counterintuitive. The importance of using videos has been emphasised countless times. Videos powerfully combine images, words and music. Several social networks had to come up with specific video features to play in this space. A commonplace suggestion for blog posts is to include videos to both engage more (video are faster/easier to go through than reading) and improve SEO performance. Videos are normally emphasised when compared with written material that requires more attention span and does not have necessarily images or photos associated with it.

When comparing videos with photos on an app, we need to take into account that access to apps is now mainly through mobile phones and possibly on the go.

However, when comparing videos with photos on an app (Instagram in this case) we need to take into account that access to apps is now mainly through mobile phones and possibly on the go. In this case, it is no longer surprising that photos get more engagement than videos; if a video is accessed for instance on public transportation, more often than not, it will be with sound off thus limiting its appeal. Also, if accessed on the go, coverage or bandwidth might be limited (think again of access while on the subway). As a consequence, a photo might work much more smoothly in this type of setting.

Furthermore, a picture is more time-efficient in conveying the message the follower likes to see as opposed to a 30-second video.


Using Celebrities? Think Twice

As far as the actual content of the post is concerned, it has been classified as product-related (P), celebrity-related (C), designer-related (D) and general-related (G).

The P type of content is about anything capturing images of actual products, may it be clothes, shoes or bags.

The C type of content pertains to pictures or videos showing celebs in different fields mainly (though not limited to) endorsing or wearing any item by the given fashion house. It is important to highlight that this category includes not only famous singers, actors/actresses, sportsmen/sportswomen, TV hosts, but also social media influencers.

The D type of content includes all the visuals that are somehow related to the designer of the eponymous fashion house. So posts showing the fashion house founder are included in this category along with any post showing the logo (e.g. the window of a recently opened shop, a close up on the logo on a piece of apparel).

The G type of content covers anything that does not fall into either of the previous categories. Among various posts, it is possible to get across posts not related to the brand or the luxury good market in general, e.g. a landscape photo in a given rural region, a photo of a famous monument or a cute pet (never underestimate a dog’s ability to drive engagement in social networks…), a sunset in a certain city, flowers or trees.

When analysing the type of content used in various posts, 72 percent of the times it was of P type, 13 percent of the times of C type, eight percent of the times of D type and seven percent of the times of G type.

In terms of engagement, the P type of content was the one among the four that performed the best. A well-taken photo closing up on a fashion item – may it be a dress, a pair of shoes or a bag – has proven to be the safest bet in terms of ability to engage followers.

C type of content on average trailed behind P type of content by seven percentage points in terms of engagement and in the majority of cases. On some occasions, engagement yielded by a celeb wearing/endorsing an item has been up to 2/3 lower than the same item worn by a standard model on an adjacent post. This being said, some posts with famous singers or actors managed to score much above the average engagement for the relevant fashion house account.

On top of that, it is worth pointing out that some posts can be further amplified by the celeb’s own account (if reposted). Also, it is worth noting that, as this category includes as well Instagram influencers, ability to engage may vary substantially according to their actual popularity. Those that are not immediately recognisable by viewers fail to impress both as celebs and as professional models.

The D type of content on average trailed the P type by 16 percent points as for engagement. From an emotional standpoint, on average this kind of posts does not convey particular messages that resonate well with followers. This being said, personal images of the founding designers yielded very high engagement especially if in private settings (e.g. pictures with family members or own pets).

The G type of content delivered lower engagement with respect to the P type by 24 percentage points. Despite investing on some occasions on high-budget videos or well-staged photos, in the vast majority of cases, the brands trying this route have not managed to get better engagement than the one yielded by product-related posts.

In this category, the posts that fared better then the average were those related to pets or messages supporting or commemorating the victims of some natural disasters or terrorist attacks.  

Finally, when splitting the subjects of posts between women’s and men’s items, the actual share of posted content reflected the actual women/men split of the analysed companies’ product range. However, when looking at the actual engagement generated, the women’s product-related posts lead men’s related-posts by eight percentage points. This could be the direct consequence of the Instagram’s audience being more skewed toward a female public.2



Based on the above results, a play-safe strategy for a fashion house on social media would play out through using, in order of priority, Instagram, Facebook, Twitter and YouTube.

As for Instagram, a brand should focus first on sharp photos of their products. This being said, it is also important to point out that engagement on a single post is of course important, but it cannot be the sole criteria to use to evaluate a campaign on.

Any follower hitting Instagram can do so at any stage of her/his customer journey; photos related to actual products can help conversion of those already deep into the sales funnel. However, more general pictures of celebs and/or the fashion house designer or logo can also help the conversion process, especially in the early stage of the customer journey generating awareness and consideration toward the brand. 

If planning to employ celebs, fashion houses had better engage real movers and shakers able to fill the engagement gap with respect to the engagement generated by product photos.   

When posting content not related to products and celebs, posting personal photos of the founder/designer can prove great at generating engagement as well.

On a more general note, fashion houses should also look at overall engagement of their followers on social media. To begin with, only a very small fraction of them is actually engaged but also, as the number of followers grows, a kind of diminishing marginal rate of return on social media activity applies (Exhibit 2). As already shown3 above, a certain threshold, followers stop being engaged as the early fans are. Mobilising these latecomers could enlarge remarkably the early part of the funnel in the conversion journey thus enhancing revenues substantially.  


About the Author

Matteo Altobelli is co-founder of Tilden/Cramm, an advisory and management company devoted to help SME’s and Startups grow their business by means of strategic analysis and management support. ( Over 20 years he has held Chief Marketing Officer positions in technology companies along with management roles in strategy consulting with Booz Allen Hamilton. He holds an MBA from INSEAD and an Aerospace Engineering degree (Magna cum Laude).


1. Bain & Co Luxury Study – Fall 2017

2. Pew Research Center – Social Media Update 2016

3. INSEAD Knowledge –


Please enter your comment!
Please enter your name here