Templates modeling just the dominant mode of gravitational radiation are generally sufficient for the unbiased parameter inference of near-equal-mass compact binary mergers. However, neglecting the subdominant/higher modes can bias the inference if the binary is significantly asymmetric, very massive, or has misaligned spins. In the first part of my talk, I will demonstrate how neglecting these subdominant modes in the parameter estimation of non-spinning binary black hole mergers can bias the inference of their population-level properties such as mass and merger redshift distributions if the detected population has a significant fraction of asymmetric mergers. Later, I will elucidate the use of subdominant modes, in addition to the dominant mode typically used in templated real-time searches, to produce significant improvements in detections and skyarea localisations in early-warning time for a range of asymmetric-mass binaries. This will enable astronomers to capture any potential electromagnetic emissions during pre/during/after merger.