Optimal Influencer Marketing Campaign Under Budget Constraints Using Frank-Wolfe

Influencer marketing has become a thriving industry with global market value expected to reach 15 billion dollars by 2022. The advertising problem that such agencies face is the following: given monetary budget find set of appropriate influencers can create and publish posts various types (e.g. text, image, video) for promotion target product. campaign's objective maximize across one or multiple online social platforms some impact metric interest, e.g. number impressions, sales (ROI), audience reach. In this work, we present an original continuous formulation budgeted influencer as convex program. We further propose efficient iterative algorithm based on Frank-Wolfe method, converges optimum low computational complexity. also suggest simpler near-optimal rule thumb, which perform well in many practical scenarios. test our heuristic against several alternatives from optimization literature standard seed selection methods validate superior performance execution time memory, its capability scale problems very large (millions) users.