Microplastic pollution is globally widespread; however, the presence of microplastics in soil systems is poorly understood, due to the complexity of soils and a lack of standardised extraction methods. Two commonly used extraction methods were optimised and compared for the extraction of low-density (polyethylene (PE)) and high-density microplastics (polyethylene (PET)), olive-oil-based extraction, and density separation with zinc chloride (ZnCl2). Comparable recoveries in a low-organic-matter matrix (soil; most >98%) were observed, but in a high-organic-matter matrix (compost), density separation yielded higher recoveries (98 ± 4% vs. 80 ± 11%). Density separation was further tested for the extraction of five microplastic polymers spiked at different concentrations. Recoveries were >93% for both soil and compost, with no differences between matrices and individual polymers. Reduction in levels of organic matter in compost was tested before and after extraction, as well as combined. Double oxidation (Fenton’s reagent and 1 M NaOH) exhibited the highest reduction in organic matter. Extracted microplastic polymers were further identified via headspace solid-phase microextraction–gas chromatography–mass spectrometry (HS-SPME–GC–MS). This method has shown the potential for descriptive quantification of microplastic polymers. A linear relationship between the number of particles and the signal response was demonstrated for PET, polystyrene (PS), polyvinyl chloride (PVC), and PE (R² > 0.98 in alluvial soil, and R² > 0.80 in compost). The extraction and identification methods were demonstrated on an environmental sample of municipal biowaste compost, with the recovery of 36 ± 9 microplastic particles per 10 g of compost, and the detection of PS and PP.