What really works for teenagers: human or fictional celebrity?

Purpose – This study aims to explore the relative effectiveness of a human celebrity endorser vis‐à‐vis a fictional celebrity or character endorser on teenage consumers' attitudes. Further, the study also seeks to assess whether the effectiveness varies depending on the nature of the product being endorsed. Design/methodology/approach – Given the purpose of the study, experimental design was used as the research methodology. In an experimental set‐up three product categories (low‐involvement food/low‐involvement non‐food/high‐involvement) and two endorsers (human celebrity/fictional celebrity) and a control group were deployed in a 3×3 full factorial design on 378 teenagers. Fictitious advertisements were used as stimuli. Findings – The study suggests that, for food and non‐food low‐involvement product categories, the impact of a human celebrity is more than that of a fictional celebrity. Regarding the purchase intentions of teenagers, it was found that a human celebrity is more effective than a fictional celebrity in food and non‐food low‐involvement products. In the case of the high‐involvement product, the human celebrity was not found to create favorable consumer attitudes. Research limitations/implications – The study results suggest that celebrity endorsements are useful, but the nature of the product also has an influence on success. One limitation of the study was the restriction to print advertisements. Practical implications – A major implication from the findings for the managers is that a human celebrity may not always be the right choice for any product promotion for teenagers. More specifically, for high‐involvement products, celebrity endorsement needs to be handled with caution since it may not prove to be successful. Originality/value – The contribution of the study is in addressing an area that has not been very well researched as yet, and in addressing a research question that has not been investigated properly.