We study the determinants of the takeover processes duration. Risk averse bidders submit bids to targets. Targets either accept, and the transaction is completed, or negotiate one more period. As time goes on, bidders and tar-gets learn about true synergies thanks to the due diligence process. But rival bidders can show up and compete to acquire targets, a desirable event from targets point of view, but costly for bidders. Using a discrete time finite horizon dynamic programming approach, our simulations characterize the relation be-tween negotiation duration, pressure of potential competition and the learning process. Our empirical exercise is based on a large sample of merger negoti-ations identified through the manual examination of SEC filings. We use the simulated method of moments to match the frequency distribution of private negotiation duration in a calibration exercise. Our results show that a 10%ex-ante probability of new bidders entering in the M&A process each month is consistent with the data.