ABSTRACT This paper summarizes and discusses the key findings of the evaluation of the English  smoking  treatment  services,  which  were  established  in  1999  as part of the English National Health Service. Within 4 years these services existed throughout the country and were working at full capacity, a total of £76 million having been spent on them over this period, excluding medication costs. In the fourth year almost 235 000 people attended treatment and set a quit date, and the total budget, including medications, was approximately £50 million. At the end of the fourth year the government allocated £138 million for the services for the period April 2003–March 2006. The CO‐validated 4‐week abstinence rate was 53%, the validated 52‐week abstinence rate was 15%, and the relapse rate from 4 to 52 weeks was 75%. There was no sex difference in cessation rates at long‐term follow‐up. The cessation results and relapse rate from weeks 4 to 52 are consistent with results from published studies, including clinical trials. The estimated cost per life‐year saved was £684 and the figure is even lower if the potential future health care cost savings are taken into account at £438 per life‐year saved. This compares with the benchmark of £20 000 per life‐year saved, which the National Institute for Clinical Excellence (NICE) is using to recommend new health care interventions in the National Health Service. The services were also succeeding in reaching disadvantaged smokers. However, there have been problems, and other health care systems considering an initiative of this kind should: set national training standards and increase training capacity before launching the services; standardize the provision of pharmaceutical treatments and make them as accessible as possible before launching the services; and give the services at least 5 years of central funding to allow them to become well established. Monitoring is extremely important but should not be so much of a burden that it detracts from developing a quality service and although cessation targets can be helpful, care needs to be taken that they are reasonable and do not promote throughput at the expense of quality.