There is accumulating evidence that the natural history of diabetic eye disease is sufficiently slow that 2-yearly retinal screening, or even longer, may be safe for some patients with diabetes. The information technology underpinning call–recall systems within screening programmes permits a more sophisticated approach to organizing retinal screening, as directed by the clinical evidence. This commentary explores the evidence for moving towards a biennial retinal screening programme for patients with Type 2 diabetes and diabetes duration of less than 10 years. Such an approach may allow capacity to introduce 6-monthly screening for high-risk patients, a targeted approach to recurrent defaulters and possible introduction of new aspects of screening such as optical coherence tomography, in addition to accommodating for the expanding number of patients with diabetes. A UK–four nations group is now critically looking at the evidence for any such changes.