Mortars are commonly used as cement-based materials, releasing substantial amounts of CO2 into the atmosphere, and the excessive use of natural resources poses a serious environmental challenge in the modern era. Therefore, the application of Supplementary Cementing Materials (SCMs) combined with a life cycle analysis (LCA) is seen as a viable approach for this dilemma. However, implementing experimental studies necessitates considerable time, specialized equipment, and dedicated space for testing and storage. In this context, Machine Learning (ML) techniques are increasingly popular for developing robust prediction models to forecast the properties of cementitious compounds containing SCMs. This study scrutinized the use of ML techniques and LCA in mortar integrating SCMs. Early in 2020s, research focus on SCMs-based mortar was limited (around 70 publications), but it grew significantly after 2023 (to about 150), with similar attention given to applying LCA and ML in mortar development (each around 25%). It was found that binary RHA utilization exhibits lower strength properties compared to other SCMs, emphasizing the need to improve RHA utilization with ternary mixtures to develop eco-friendly, durable, and cost-effective mortar by augmenting RHA fractions. Additionally, a ternary mortar mixture with a calcium source proved the effectiveness of this approach. However, further investigation is needed into reducing cement content beyond 40%. Approximately 60% of studies focused on predicting compressive strength, while limited focus was given to tensile strength, flexural strength, yield stress, and drying shrinkage. Recently, a wide variety of machine learning techniques has been employed in the property predictions of mortar, with an escalating emphasis on tree-based ensemble methods such as bagging and boosting techniques (more than 50% publications) rather than depending on single learner models. However, a limited use of hybrid techniques (6% of published literature) was observed in the property predictions.