A descriptive analysis of product incentivization in popular YouTube skincare videos

YouTube's $532 billion beauty industry serves as a lucrative, efficient marketing platform for retailers, as advertisements are often presented to encourage viewers to purchase and use skincare products.1 Incentivization, defined as the practice of attaching rewards to motivate certain behaviors, is common on YouTube and is present in three major forms: (1) explicit sponsorship whereby the uploader is paid by the sponsoring company; (2) affiliated links or coupon codes whereby purchases made by viewers through the link/code earn the uploader a commission; and (3) free products sent by companies to uploaders with hope of exposure.2 Our objective was to describe the incentivization of skincare products on YouTube. We conducted a YouTube search for “skincare routine,” analyzing the 100 most-viewed, relevant (i.e., anti-acne, anti-aging, moisturizing) English-language videos. Products advertised within each video were characterized by marketing status (incentivized vs. non-incentivized) and brand type (e.g., high-end, drugstore). Each product's ingredient list was reviewed and potential skin allergens were recorded.3 Number of subscribers per uploader at time of video upload was obtained from web.archive.org. The primary outcome was product incentivization, defined by explicitly disclosed sponsorship, or presence of an affiliated link and/or coupon code; prevalence of companies sending products to uploaders was unable to be assessed. Generalized estimating equations was applied to model the number of products advertised per video and calculate odds of product incentivization. We compared incentivized to non-incentivized videos, estimating the difference in number of products marketed per video using least-square means. Analyses were performed in SAS 9.4. The 100 most-viewed YouTube videos had over 139 million combined views over 9 years, with the top 20 videos containing 65% of all views. Nonmedical individuals (79%) uploaded the majority of videos (Table 1). The majority of products were incentivized (67%) and high-end (45%), with affiliate links included (44%). View count did not increase the likelihood of product incentivization (OR 0.92, 95%CI 0.60–1.40). The number of subscribers of video uploaders increased the likelihood of video incentivization (OR 1.60, 95%CI 1.20–2.13). Videos with incentivized products promoted a mean of 9.33 products per video compared to 8.39 products per video for non-incentivized products (p = 0.23). Additionally, increasing video incentivization did not increase the likelihood of skin allergens being present within ingredient lists of featured products (OR 1.22, 95%CI 0.70–2.11). In both high-end and drugstore products, phenoxyethanol was the most common skin allergen, followed by fragrance/parfum. The skewed distribution of views toward the top 20 videos suggests that there may be a positive-feedback-loop in which viewers prefer more popular videos. However, we found that subscriber count rather than view count predicted product incentivization. This is likely because uploaders with more subscribers (i.e., social media “influencers”) receive more brand deals and sponsorship offers from companies, as they have greater potential to reach a wider audience. Products advertised by high subscriber count content creators were also mostly high-end, and thus likely less affordable and accessible than drugstore products. Of note, incentivized products are also not necessarily those a board-certified dermatologist would recommend. Limitations of this study included videos not directly disclosing the presence or absence of sponsored products. As social media platforms become increasingly popular for sharing medical information, we encourage dermatologists to engage in conversations with their patients regarding dermatology-related content they have seen online.4, 5 Additionally, future behavioral economics research should investigate how different methods for framing incentives (e.g., implementation intention, in which viewers are prompted to state a plan for how they might use a product) ultimately influence consumer behavior.6 Due to the widespread incentivization of skincare products on YouTube, we urge viewers to exercise caution when considering these videos as a source of health information. This study met the definition of Institutional Review Board exempt research. LX wrote the manuscript and was responsible for study ideation. NLT, AW, SR, and AM assisted with data collection. CRC provided all statistical analyses. DXZ, CW, and JFS provided administrative support. All authors reviewed and approved the final manuscript. LX and JFS take responsibility for the paper as a whole. The authors report no conflicts of interest relevant to this work. This study met the definition of Institutional Review Board exempt research, as it utilized publicly available data from YouTube that was deidentified on aggregate presentation.