I study a staggered policy change directly affecting only the asking price of homes for sale, using a novel and large micro dataset. I find that increasing the asking price reduces the number of online views, the number of visitors to the open house viewings, the number of bidders participating in the auction-like sales process and increases the time on market. Contrary to predictions of standard theory I find no effect on the sales price, nor on the sales effort exerted by real estate agents. To explain these findings I develop a simple search model where the non-committing asking price directs search and real estate agents set the asking price optimally. The model is consistent with the empirical results when home buyers follow a simple rule of thumb, but not when they act fully rational.