Since its advent 17 years ago, stimulated Raman scattering (SRS) microscopy has emerged as a transformative imaging modality by visualizing chemical bonds with high sensitivity, speed, specificity, and resolution. Despite its enormous success, a rigorous theory is yet lacking in the community. The fundamental question of just why and how much SRS microscopy can outperform conventional Raman microscopy has not been quantitatively answered. Raman scattering, traditionally understood through the Raman cross section (σ), has long been believed weak due to its extremely small values when compared to linear absorption cross sections. However, this view is incomplete and even misleading since Raman scattering and linear absorption processes involve different orders of light-matter interaction. In this review, we summarize the recently developed stimulated response formulation, which defines the stimulated Raman cross section (σ) in the same spirit of Einstein's B coefficient. Unlike σ, σ turns out to be intrinsically strong and even exceeding the electronic counterparts, which is supported by experimental measurements and quantum electrodynamic theories. This new framework reveals a previously unknown duality nature of Raman scattering, where both σ and σ can exhibit vastly different magnitudes for the same molecule, connected by the influence of vacuum zero-point fluctuations. Additionally, the Raman duality also generalizes Einstein's coefficients so that four processes (spontaneous and stimulated emission, spontaneous Raman and SRS) are unified. Finally, the formulation provides quantitative prediction of the absolute signal and detectability of SRS microscopy. We can prove that SRS excels in high spatiotemporal regimes, explaining its unparalleled ability to image chemical bonds, which inherently demand high spatial and temporal resolution. We expect this theory to facilitate both scientific understanding and technological applications of Raman spectroscopy.