A first-degree relative of an individual with colorectal cancer is on average at about a twofold increased risk. This could not occur without there being strong underlying risk factors that are correlated in relatives. About 90% of colorectal cases occur in people who are above median familial/genetic risk, so there is great potential to use genetics to prevent colorectal cancer. Two rare inherited syndromes have been identified: familial adenomatous polyposis (FAP) and hereditary non-polyposis colorectal cancer (HNPCC). The former appears to be mostly due to mutations in the APC gene, and the latter to mutations in mismatch repair (MMR) genes, so it would be better named as hereditary mismatch repair deficiency (HMRDS). By fully characterising a population-based series of early-onset cases, we have shown that MMR gene mutation carriers and their relatives can be more efficiently identified by characterising the tumours of early-onset cases, independently of their cancer family history, using immunohistochemistry (IHC)-not microsatellite instability (MSI) testing. This identifies the specific MMR gene likely to be involved, reducing the costs of mutation testing. Identification of genetically susceptible individuals using the tumour phenotype of affecteds, rather than family cancer history, could become the standard approach of cancer genetic services in the twenty-first century, and could lead to cancer prevention in individuals who are at a high genetic risk when young. There is an urgent need for research on the efficacy and optimisation of surveillance procedures in these high-risk individuals, and identification of the environmental, lifestyle and other genetic factors that exacerbate, or ameliorate, risk in mutation carriers.