The choice of biomass and volume models is one of the main sources of uncertainty in forest biomass assessment. The quality of the dataset used to develop the equations, the statistical procedures used for modelling and the accuracy in reporting the results vary greatly among the models. Recent studies highlighted that about 25% of the published equations contain errors, omissions or predict unrealistic values. Considering that the number of published models is constantly growing, the lack of consistent methods can make model selection more difficult and considerably affect the accuracy of the estimates. Before choosing and applying the models, it would be therefore necessary to perform transparent quality control and verification procedures in order to establish their degree of reliability. This paper presents a method for assessing allometric equations for quality by providing transparent routines and consistent checks, and, at the same time, applicable to a large number of equations. These procedures are meant to quantify the degree of confidence of the tree biomass allometric equations available in the literature and their potential to be used for forest carbon stock assessment. Consistency and integrity of models’ metadata are checked in order to identify possible errors in data input. “Key metadata”, essential to ensure the applicability and documentation of the models, such as taxonomy, units and sources, are tested. Reliability of model predictions is assessed by comparison with external data set and models, customizable by the users. A reliability rate will be provided. This is a first attempt to provide a flexible, fast and user-friendly procedure for model selection and verification that can help to identify and reduce possible errors and omissions, facilitate the review process and consequently improve transparency, consistency, comparability, completeness, and accuracy of the biomass estimates worldwide.