The ability to estimate a supplier's marginal cost confers a strategic competitive advantage to the buyer, whether in negotiations with suppliers, in an auction setting, or when an auction is used to initiate the process, which is then followed up with a traditional negotiation. Focusing on electronic reverse auctions characterized by a one‐shot, first‐price, sealed‐bid format, this article proposes an approach for estimating a supplier's marginal cost. Specifically, we suggest a two‐stage model: In the first stage, empirical analysis is used to predict the winning bid. In the second stage, a game‐theory approach is used to refine the outcome of the first stage to provide an estimate of the supplier's marginal cost. To assess the model, we apply it to data from a food and beverages company that carried out electronic auctions to select suppliers for industrial maintenance services. We find that our estimates are very close to those made by the suppliers and compare favorably to the efficient marginal costs determined with the widely used approach of data envelopment analysis. This also implies that after selecting a supplier through an auction, the buyer can enhance follow‐up negotiations with the supplier by contrasting our model's estimates of the marginal costs with the supplier's' inefficiencies detected with the data envelopment analysis.