Influencer–Brand Partnership: A Matching Approach

My job market paper, “Influencer–Brand Partnership: A Matching Approach,” develops the first empirical structural model of how brands choose influencer bundles on Instagram. Using proprietary data from a leading Iranian influencer marketing platform—spanning over 11,000 campaigns and 100,000 ads—I study how brands match with influencers across different campaign types (CPM, Reach, Influence) and content formats (Post vs. Story).

I build a structural matching model based on the pairwise stability framework of Fox and Bajari (2013), allowing me to identify brand preferences over influencer bundles without observing the full choice set. Estimation reveals systematic differences across campaign goals: for example, follower growth campaigns favor engagement and niche category targeting, while sales-oriented campaigns emphasize different metrics.

I conduct counterfactuals to assess how platform design and influencer tier restrictions affect welfare and match quality. I also extend the model to account for posted influencer prices, using a control function approach to address endogeneity.

This framework contributes broadly to the empirical marketing literature by modeling multi-partner matching in environments with large choice sets and limited transaction-level data.