Litnet

Abstract

Social media has not only become an important feature of everyday life, but also an important platform for marketing purposes. Social media allows companies to market products and engage with customers and potential customers, build their brands and develop a closer relationship with customers. Because of its utility, most companies have some form of social media presence, and one of the most popular social media platforms in particular, Instagram, has become known for its marketing potential, either for companies directly or via influencers. In South Africa, 68% of the country’s largest brands use Instagram, making Instagram the fourth most popular social media platform for marketing purposes in South Africa. In addition, Instagram’s user numbers are growing annually worldwide and also in South Africa – the number of Instagram users in South Africa has grown from 3,8 million in September 2018 to 5,15 million by August 2020. This suggests that Instagram may become an even more important marketing platform in the near future.

The current study examines the Instagram posts of three of the most important South African publishers: NB Publishers (@nbpublishers), LAPA Publishers (@lapa_uitgewers) and Protea Boekhuis (@proteaboekhuis), from their first posts to 30 April 2020. Their posts are compared to recent research in terms of the number of followers, the frequency of posts versus user responses, the format of posts versus user responses, and the use of hashtags. Finally, machine learning is used to determine the emotional content of posts and compare results with previous research on Instagram as a marketing platform. The aim is to determine how actively these publishers use Instagram and to what extent users respond to their posts, as well as to test various factors that could potentially influence user reactions.

It is found that these publishers have only recently started using Instagram and that they do not post daily, but that the majority of posts are positive posts in a photo format. These positive posts also receive more likes on average than negative posts, although the difference between the number of likes for positive and negative posts is small. However, the number of hashtags used with each post does not play a role in the number of likes that the post receives.

These publishers’ Instagram accounts differed in important ways (until 30 April 2020), as shown in the following table.

LAPA Publishers has the oldest Instagram account of these three publishers and is the most active in terms of the number of posts, but, on average, posts by LAPA receive around half the number of likes as posts by NB Publishers. LAPA also posts the largest percentage of positive posts, while Protea Boekhuis uses the largest number of hashtags per post. Yet, despite previous research suggesting that the frequency of posts and the number of hashtags used should affect user reactions, along with the sentiment of the post, the greatest determining factor for user reactions was found to be the number of followers. Given NB Publishers’ age and prestige, along with these figures from their Instagram account, this publisher’s popularity on Instagram is interpreted as a function of the prestige they have acquired as one of South Africa’s top publishers. In other words, the study finds that intrinsic aspects on Instagram are insufficient to explain user reactions in this case, and that extrinsic factors such as the prestige of the brand – as acquired outside Instagram – are important contributors to user reactions on Instagram.

Keywords: Instagram; machine learning; marketing; publishing; social media

Lees die volledige artikel in Afrikaans

’n Ontleding van drie Suid-Afrikaanse uitgewerye (NB-Uitgewers, LAPA Uitgewers en Protea Boekhuis) se Instagram-plasings