Adam Elwood Alessandro Rozza Elanor Colleoni Angelo Miglietta

Measuring Brand-Influencer Visual Congruence on Instagram Using Deep Learning and Automated Image Recognition

Are you already subscribed?
Login to check whether this content is already included on your personal or institutional subscription.

Abstract

Influencer Marketing refers to the practice of remunerating influencers to endorse a brand on social media. For this marketing practice to be successful, extant research has shown that a certain degree of congruence between the influencer image and the brand endorsed is necessary. This congruence has so far been studied as textual congruence. However, as influencer marketing is increasingly conducted via image-sharing platforms, such as Instagram, where influencers create visually appealing content to engage with their followers, we lack a way to identify and evaluate the role that visual congruence between brand and influencers play in driving the followers engagement. In this paper, we apply deep learning and image recognition algorithms on a large sample of images from Instagram, to extract a suitable measure of visual congruence as emerging from the automated analysis of brand and influencers images posted overtime. The algorithm is presented in detail along with examples of top and worst ranked brand-influencer visual congruences. To validate our measure, we run a multiple linear regression analysing the influencer post engagement levels against the brand-influencer visual congruence measure. Evidence is found of a significant impact of the brand-influencer visual congruence on the followers level of engagement on influencers posts endorsing a brand. Implications for academics and practitioners are drawn.

Keywords

  • Influencer marketing
  • social media
  • visual congruence
  • Instagram
  • deep learning
  • algorithms

Preview

Article first page

What do you think about the recent suggestion?

Trova nel catalogo di Worldcat