Personality is important for a range of life outcomes. However, despite evidence that personality changes across time, there is a concerning tendency for researchers outside of personality psychology to treat measures of personality as if they are non‐changing when establishing whether personality predicts important life outcomes. This is problematic when personality changes in response to outcomes of interest and creates a methodological issue that may result in misleading conclusions. We illustrate this methodological issue and suggest using measures before the outcome takes place to mitigate concerns. We then demonstrate, using data from Germany, that using post‐event personality measures, as opposed to pre‐outcome measures, to predict both occurrence of, and reactions to, socio‐economic events results in inconsistent conclusions in the directions hypothesized and therefore increases the likelihood of Type 1 and Type 2 errors. This has implications for research investigating the importance of personality for psychological, behavioural, and socio‐economic outcomes. Copyright © 2017 European Association of Personality Psychology