We introduce a new method for retrieving formaldehyde (HCHO) based on principal component analysis (PCA) of satellite-measured radiances. Applying the technique to the Suomi National Polar-orbiting Partnership/Ozone Mapping and Profiler Suite (S-NPP/OMPS) radiances between 328.5 and 356.5 nm, we extract principal components (PCs) associated with various physical processes (e.g., ozone absorption and rotational Raman scattering) and measurement details (e.g., wavelength shift). These PCs, along with precomputed HCHO Jacobians, are utilized in spectral fitting to estimate HCHO loading and reduce interferences. Comparisons with model simulations and independent Ozone Monitoring Instrument (OMI) retrievals indicate that our algorithm can detect enhanced HCHO signals over source regions such as the southeast U.S., producing HCHO total columns with similar spatial distributions and seasonal patterns. While our OMPS retrievals are ~15–20% lower than OMI retrievals from a different algorithm, the differences may be attributed to several instrumental and algorithmic factors. This study demonstrates the potential of PCA algorithms and of OMPS for continuing the long-term satellite HCHO data record.