Biogeochemical process rates are often estimated by quantifying the rate of change in a reactant over space or time. Nutrient spiraling estimates quantify uptake length as the inverse of the change in concentration over distance. Denitrification and nitrogen fixation rates can be estimated as net N2 fluxes based on changes in N2 concentration over time. Decomposition rates are quantified as mass-loss over time. Ultimately, each of these approaches quantify a decay rate, typically referred to using k, which can then be converted to process rate estimates by scaling up temporally or areally. When process rates are high, and field and analytical approaches are precise, these approaches are fairly straightforward. However, quantifying decay rates is often not as straightforward as it seems, and commonly involves lower process rates with outlier data points due to either contamination, lack of analytical precision, or demonic intrusion. Deciding which data points to include, which to drop, and how to decide whether a dataset provides a quantifiable decay rate or not can be challenging. Individual choices on whether to keep or remove an individual point can have a considerable effect on the ultimate process rate estimate, but these decisions are rarely described or scrutinized in any transparent way. For this presentation, we will present examples of “messy data” from real-world examples where decisions on whether to keep or remove individual data points effect the ultimate outcome. We will present what we consider to be best practices and strategies to reduce any individual bias from this approach, and will present an opportunity for individuals to become involved in a project to quantify how the art of biogeochemistry influences our ultimate process rate estimates. Ultimately, with this presentation, we aim to move towards a more unified approach to quantifying decay rates to allow for more repeatable, transparent science.