This paper presents a summary of methodologies previously developed that focus on dealing with no historic degradation data, with changes in degradation condition and incorporating statistical monitoring into conditionbased maintenance (CBM). However, estimating the remaining useful life (RUL) of an asset is a complex task widely studied in the last decades. The main difficulty in RUL estimation through condition monitoring is the variety of failure paths that the assets develop. Studies have shown that similar assets develop significantly different degradations paths affecting the amount of historic data available for its analysis. Consequently, the available historic data to study these degradations are even scarcer. In addition, the condition variable might suffer a change through the asset life, producing misleading estimations if these changes are not taken into account when fitting a model. By considering these changes and highlighting the obstacles faced in the development of these methodologies, the lessons learned and the challenges in future RUL estimation research are further explored.