We consider the problem of how to optimally close a large asset position in a market with a linear temporary price impact. We take the perspective of an agent with a market opinion that translates into a (linear) drift in asset price dynamics. By appealing to classical stochastic control we derive explicit formulas for the closing strategy that minimizes a sum of execution costs and a quadratic risk functional. We then proceed by comparing agents observing a signal about the asset's future price with agents who do not see the signal. We compute explicitly the expected additional gain due to the signal, and perform a comparative statics analysis