In a recent paper, Gerritsen et al. (2022) propose to modify the well-accepted sampling strategy in paleomagnetism by collecting more single-sample sites. They also argue that the paleomagnetic community commonly applies a loosely defined set of quantitative data filters and that there is no need for an expert-eye to analyze and interpret paleomagnetic data. Many paleomagnetists will disagree with these claims as paleomagnetic methods provide very robust results at the site level when the study is done with sufficient rigor. As stated in Gerritsen et al. (2022) they deliberately kept directions that an experienced paleomagnetist would likely immediately discard as unreliable. Can we really draw conclusions from such an approach to paleomagnetism? The strategy proposed by Gerritsen et al. (2022) has serious drawbacks well illustrated by the datasets from Turkey (van Hinsbergen et al., 2010), Mongolia (van Hinsbergen et al., 2008), Norway (Haldan et al., 2014), and Antarctica (Asefaw et al., 2021) used by Gerritsen et al. (2022). The main objective of this comment is to support standard methods (Butler, 1992; Tauxe et al., 2018) for a well-defined determination of the paleomagnetic direction per site based on the sampling of several samples per site.