Screening exfoliative cytology for early dysplastic cells reduces incidence and mortality from squamous carcinoma of the cervix. In the developed world, screening programmes have adopted a 3-5 years recall system. In its absence, cervical cancer would be the second most common female cancer in these regions; instead, it is currently eleventh. However, there exist a number of limitations to the smear test even given the removal of contaminants using liquid-based cytology. It is prohibitively expensive, labour-intensive and subject to inaccuracies that give rise to significant numbers of false negatives. There remains a need for novel approaches to allow efficient and objective interrogation of exfoliative cytology. Methods that variously exploit infrared (IR) microspectroscopy are one possibility. Using IR microspectroscopy, an integrated 'biochemical-cell fingerprint' of the lipid, protein and carbohydrate composition of a biomolecular entity may be derived in the form of a spectrum via vibrational transitions of individual chemical bonds. Powerful statistical approaches (e.g. principal component analysis) now facilitate the interrogation of large amounts of spectroscopic data to allow the extraction of what may be small but extremely significant biomarker differences between disease-free and pre-malignant or malignant samples. An increasing wealth of literature points to the ability of IR microspectroscopy to allow the segregation of cells based on their disease status. We review the current evidence supporting its diagnostic potential in cancer biology.