The Journal of Smoking Cessation 6(1): 3-8 Abstract Smoking prevalence, at any point in time, accounts for the balance of smokers ending the habit and people starting or returning to smoking. Very different scenarios can result in the same prevalence, and each can have different implications for health services. Therefore understanding the degree of churn in the smoking population is essential. This is an area where there is little or no data. In such circumstances, Delphi methods utilising expert opinion can provide robust forecasts for planning purposes. A Delphi survey, involving 300+ general practitioners (GPs) in Northern England, projected a modest 2.2% reduction in smoking prevalence over 10 years, but this involved a reduction of nearly 30% in the existing cohort of smokers, countered by new/returning smokers equivalent to 21% of the cohort. Different age groups showed different patterns. The findings suggest that approximately 1 in 6 smokers need to set a quit date each and every year to achieve even a 2.2% reduction. Currently, 1 in 16 smokers set a quit date each year. Although `early wins’ may have been achieved, the challenges to achieve continuing reductions in smoking prevalence are considerable. Preventing young people starting to smoke is key.