We rely on novel textual analysis of real estate listings and identify renovated dwellings in a data set of Norwegian transactions to estimate the renovation premium in an urban housing market. The renovation premium is estimated by classical regression approaches as well as random forests models. The strength of the latter is that it allows for a more complex interplay between the renovation premium and explanatory variables. We find a significant positive renovation premium of 5-7 percent for renovated dwellings and a negative premium of 9-10 percent for unmaintained/neglected dwellings. These averages mask significant variation in these premiums over time. In particular, there is a counter-cyclical effect. In a hedonic price model, omitting renovation has implications for estimated short-term house price growth. We also find that unmaintained dwellings tend to transact more in the fourth quarter, indicating that parts of seasonal price variation reported in the literature are due to compositional variation with respect to renovation. This composition effect tends to bias price movement estimates downward, if uncontrolled for, as unmaintained dwellings tend to transact at a significantly lower price.