Problematic smartphone use is an emerging issue in behavioural addiction research. At the same time, measuring smartphone use with mobile apps has become increasingly common. However, understanding how much data is necessary requires careful consideration if the field is to move forward. Here, we examine how much time should be spent measuring mobile phone operation in order to reliably infer general patterns of usage and repetitive checking behaviours. In a second analysis, we consider whether a self-report measure of problematic smartphone use is associated with real-time patterns of use. Results suggest that smartphone usage collected for a minimum of five days will reflect typical weekly usage (in hours), but habitual checking behaviours (uses lasting less than 15 seconds) can be reliably inferred within two days. These measurements did not reliably correlate with a self-reported measure. We conclude that patterns of smartphone use are repetitive and our results suggest that checking behaviour is a particularly consistent and efficient measure when quantifying typical and problematic smartphone usage.