This study aims to quantitatively explore the multifaceted determinants that influence earnings in the translation industry. Using a dataset comprising 45 000 translator profiles, the study focusses on delineating disparities correlated with demographic variables such as age, gender, home country wealth, and language pairs. The study uses a random forest regression model to delineate the complex interaction between gender, the economic standing of a translator's domicile country, age, and linguistic proficiency, as they relate to earnings. Our findings substantiate and, in many ways, extend existing qualitative and anecdotal evidence that has shaped the discourse in this sector. The rigorous empirical framework employed here can be replicated or adapted to study other sectors within the gig economy, thus contributing to a more comprehensive understanding of labour dynamics in the digital age.