Paper Number



This study empirically analyses and quantifies the impact of advertising disclosures on ad effectiveness based on a large real-world dataset comprising of Instagram posts from actual influencers. We utilize that enforcement differs by region and use data from regions with proper disclosure-culture to train a stochastic gradient boosting model that predicts if a post is advertising. By using model predictions for a large sample of >34.000 posts from regions other than the training set, we aim to tell apart the effect of disclosure from the effect of a posts' inherent advertising nature. We find that at least half of the penalty associated with disclosures may be attributed to the posts' advertising nature. Further, we identify negative spillover effects for performing undisclosed advertising on other non-advertising posts of the same influencer. Spillover effects are not detected for disclosed advertising. We conclude that the downside of undisclosed advertising may outweigh its upside.



When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.