IDLab Seminar: Purchase Drivers in the Online Resale Market
On September 25, 2026, the International Laboratory of Intangible-driven Economy (IDLab) held the first research seminar of the academic year. Research Fellow Elena Zelenskaya and Research Assistant Olga Okulova presented the findings of their study on which seller signals shape consumer decisions in the online resale market, focusing on the luxury jewelry segment.
The luxury resale market is growing rapidly; however, online platforms face pronounced information asymmetry, as buyers cannot physically inspect items, increasing the risks of purchasing counterfeit or misrepresented goods. Focusing on women's jewelry, the authors addressed several key gaps in the literature. In particular, the researchers shifted the focus from consumer motivations to seller characteristics and accounted for the specifics of a distinct product category. Moreover, rather than relying on primary data collection, the study utilized data from a major Russian resale platform covering 3,000 sellers and 12,000 listings. The number of likes—reflecting items added to wishlists—served as a proxy for purchase intention. This approach opens up opportunities for analyzing open platform data, even though the choice of such a proxy remains a subject of academic debate.
The findings yielded unexpected results. Contrary to earlier literature, the hypothesis that professional sellers inspire greater trust was not supported. Instead, hard-to-imitate signals, such as verified accounts and a long tenure on the platform, exerted a statistically significant positive effect on buyer interest. At the same time, the number of followers had a positive but modest effect. The practical value of the study suggests that resale platforms should shift their focus from growing seller follower counts to promoting verification tools designed to mitigate authenticity-related risks.
During the discussion, seminar participants addressed methodological questions raised during the peer-review process of the paper. A central topic was justifying the use of likes as a proxy for purchase intention. In particular, suggestions included refining the conceptual model specification and exploring alternative specifications, such as normalizing likes by the seller's follower count and incorporating listing duration as an additional metric. Furthermore, participants highlighted the asymmetric sample structure dominated by individual sellers and recommended testing interaction effects between seller type and account verification status.
