Municipal solid waste (MSW) generation forecasting has become an important tool for decision-making in urban environments, not only due to its essential role in effective waste management, but also because it provides an understanding of the complexity of the factors that govern it. Current research bases its forecast models (e.g., artificial neural networks, regression methods, three decision methods…) on predictive variables supported by pre-existing government information or, alternatively, on related studies with different site characteristics due to the lack of primary data from the specific sector. These assumptions and generalizations generate a different representation of the area of interest, raising the level of uncertainty of the results and reducing their level of reliability. The current review focuses on exploring the influence, relevance and opportunities for improvement when it comes to including or excluding sociocultural, environmental and/or economic variables in the solid waste forecasting process. Relevant information has been provided regarding the predictor variables considered to have better predictive power and, at the same time, limitations in data availability have been highlighted. Finally, it is concluded that the adoption of case study-specific predictor variables collected through primary data (e.g., questionnaires or surveys) would improve the predictive performance of the models providing a robust and effective tool for waste management. In addition, it is expected that the recommendations provided will be useful for future research related to MSW prediction and, thus, contribute to obtaining more representative results.