This article report evidences from an economic experiment on two-dimensional first-score procurement auctions, where subjects bid on both price (financial score) and quality (technical score), and bids are evaluated by a scoring rule designed by a buyer. The experimental design consists of two between-subject treatments of interest: the first in which quality matters more on the final score and the latter in which the price bid is instead determinant on the final result. To analyze our empirical findings, we tested them against the theoretical benchmark, which corresponds to the standard Bayesian-Nash equilibrium of the underlying game. Moreover, we are interested in analyzing how the bidding behavior changes when the final score is determined by these different criteria and the effects in terms of efficiency. We found that the theoretical model succeeds in predicting the subject’s choice of quality but not the choice of price. Moreover, defining the bidder with the highest potential score as the most “efficient type”, we observe that, when the final score in mainly determined by the quality, there is an higher frequency of efficient winner than the case in which is assigned an higher weight to the financial score.