This review updates the molecular pathological classification of colorectal cancer (CRC), focusing on recent advancements in bulk and single-cell RNA data analysis. CRC is characterized by two major mechanisms of genomic instability: chromosomal instability (CIN) and microsatellite instability (MSI). The consensus molecular subtypes (CMS) classification, based on bulk RNA sequence data, identifies four major groups. However, inter-tumoural and intra-tumoural heterogeneity remains significant, leading to the development of more refined classifications such as the Colorectal Intrinsic Subtypes (CRIS) and the Pathway-Derived Subtypes (PDS). The CRIS classification removes the influence of the tumour microenvironment, while the PDS classification uses Gene Ontology inferred pathway activation patterns to reveal a continuum of intrinsic biology associated with biological, stem cell, histopathological, and clinical attributes. Recent studies have also explored the role of the stromal and immune components in CRC, highlighting their importance in prognosis. Single-cell transcriptomics has provided additional insights into the transcriptomic dichotomy of malignant cells, leading to the development of the single-cell intrinsic CMS (iCMS) classifier. This classifier identifies two distinct epithelial classes with distinct gene expression, transcriptional factor activity, and genomic profiles. The review also discusses the limitations of gene-based classifiers and the advantages of pathway-based classifiers, which provide a more reproducible way to test associations between samples. The pathway-derived subtypes (PDS) classification, based on Gene Ontology signatures, reveals granular information within the largest tumour subtype defined by the CMS classification. Looking forward, the review emphasizes the importance of pathology in guiding molecular subtyping discoveries, particularly in the era of spatial profiling technologies. These technologies offer the potential to reveal detailed insights into the intricate mechanisms underlying cancer development and progression, providing a more comprehensive understanding of the complex signaling pathways involved.This review updates the molecular pathological classification of colorectal cancer (CRC), focusing on recent advancements in bulk and single-cell RNA data analysis. CRC is characterized by two major mechanisms of genomic instability: chromosomal instability (CIN) and microsatellite instability (MSI). The consensus molecular subtypes (CMS) classification, based on bulk RNA sequence data, identifies four major groups. However, inter-tumoural and intra-tumoural heterogeneity remains significant, leading to the development of more refined classifications such as the Colorectal Intrinsic Subtypes (CRIS) and the Pathway-Derived Subtypes (PDS). The CRIS classification removes the influence of the tumour microenvironment, while the PDS classification uses Gene Ontology inferred pathway activation patterns to reveal a continuum of intrinsic biology associated with biological, stem cell, histopathological, and clinical attributes. Recent studies have also explored the role of the stromal and immune components in CRC, highlighting their importance in prognosis. Single-cell transcriptomics has provided additional insights into the transcriptomic dichotomy of malignant cells, leading to the development of the single-cell intrinsic CMS (iCMS) classifier. This classifier identifies two distinct epithelial classes with distinct gene expression, transcriptional factor activity, and genomic profiles. The review also discusses the limitations of gene-based classifiers and the advantages of pathway-based classifiers, which provide a more reproducible way to test associations between samples. The pathway-derived subtypes (PDS) classification, based on Gene Ontology signatures, reveals granular information within the largest tumour subtype defined by the CMS classification. Looking forward, the review emphasizes the importance of pathology in guiding molecular subtyping discoveries, particularly in the era of spatial profiling technologies. These technologies offer the potential to reveal detailed insights into the intricate mechanisms underlying cancer development and progression, providing a more comprehensive understanding of the complex signaling pathways involved.