Estimates of illegal wildlife trade vary significantly and are often based on incomplete datasets, inferences from CITES permits or customs seizures. As a result, annual global estimates of illegal wildlife trade can vary by several billions of US dollars. Translating these figures into species extraction rates is equally challenging, and estimating illegal take accurately is not achievable for many species. Due to their nesting strategies that allow for census data collection, sea turtles offer an exception. On the Caribbean coast of Costa Rica, three sea turtle species (leatherback,Dermochelys coriacea;green,Chelonia mydas; and hawksbill,Eretmochelys imbricata) are exploited by poachers. Despite the consumption of turtle eggs and meat being illegal, they are consumed as a cultural food source and seasonal treat. Conservation programmes monitor nesting beaches, collect abundance data and record poaching events. Despite the availability of robust long-term datasets, quantifying the rate of poaching has yet to be undertaken. Using data from the globally important nesting beach, Tortuguero, as well as beaches Playa Norte and Pacuare on the Caribbean coast of Costa Rica, we modelled the spatial and temporal distribution of poaching of the three sea turtle species. Here, we present data from 2006 to 2019 on a stretch of coastline covering c.37 km. We identified poaching hotspots that correlated with populated areas. While the poaching hotspots persisted over time, we found poaching is declining at each of our sites. However, we urge caution when interpreting this result as the impact of poaching varies between species. Given their low abundance on these beaches, the poaching pressure on leatherback and hawksbill turtles is far greater than the impact on the abundant green turtles. We attribute the decline in poaching to supply-side conservation interventions in place at these beaches. Finally, we highlight the value of data sharing and collaborations between conservation NGOs.