How high should the first offer be? Prior to any negotiation, decision-makers must balance the tradeoff between two opposing first-offer effects. On the one hand, more assertive first offers benefit negotiators by anchoring the negotiation in their favor. On the other hand, a first offer that is too assertive increases impasse risk. Past research has demonstrated either the first offer’s anchoring benefits (while largely ignoring the risk of impasse) or its impasse risk (while largely ignoring anchoring benefits). The literature also frequently builds on simulated laboratory or classroom scenarios and has yet to provide an empirical, applied answer to the question of how high the ideal first offer should be. We integrate these separate literature streams and establish, based on over 25 million incentivized real-world sales negotiations,(1) a linear anchoring effect of first offers on sale prices and (2) a nonlinear quartic effect on impasse prevalence. We further identify three magnitude zones with distinct first-offer effects, identify specific points with particularly low impasse risks and high anchoring benefits, empirically examine the opening-offer midpoint bias—the assumption that buyer and seller eventually meet in the middle of their opening offers—and establish moderation by price certainty and product demand (the impasse risk decreases, the more uncertain a product’s objective value is and the fewer potential buyers are interested). Finally, we apply machine learning analyses to predict agreements and impasses and present a website that provides first-offer advice configurable to negotiators’ particular product, list price, and risk preferences.