In the Al Shaheen (ALS) field offshore Qatar, injection wells have generally been back-flowed to remove solids introduced during drilling and stimulation, with the objective of enhancing injectivity. This is considered by many as "best-practice" in the industry. Data from the early back-flow period for injectors and the early production period for producers has been reviewed to try to identify "clean-up" events. Some apparently spontaneous rate increases were observed. Analysis of this data was found to be complicated because during the initial 20 days of production frequent choke size changes were made. Bottom-hole pressure data around the time the wells were opened was also found to be absent in most cases. An attempt to compare the injectivity of a few wells which were not back-flowed with analogues that were was also frustrated, due to variability of permeability and oil viscosity between wells. It was not possible to draw definitive conclusions. A new phase of drilling at ALS provides an opportunity to investigate the efficacy of back-flowing in a pre-mediated fashion. Special provisions will be made during the initial production period to evaluate whether or not the wells clean up. In addition a comparison will be made between the injection performance of pairs of injection wells which are located in areas with similar transport properties. For the injection well pairs, one injector will be back-flowed for a month while the other will be put on injection immediately after stimulation. In this manner it is hoped to demonstrate conclusively whether back-flow of ALS injectors enhances injectivity. Introduction Historically, most Al Shaheen (ALS) water injection wells have been back-flowed before conversion to water injection. Back-flow was done to "clean-up" the injectors and also to produce oil revenue. This paper reviews the possible benefits of back-flow, based on experience with existing wells, and describes steps that will be taken in future to determine the effect of back-flow in a few wells in a carefully controlled environment. The wells in this future trial of back-flowing will each have a "control" well that will not be back-flowed before water injection starts. Although many ALS wells showed apparent "clean-up" events in their early production, it is by no means certain that there was a lasting improvement. The results of back-flow were hard to analyse because, during the early days of back-flow when "clean-up" events were expected to occur, the choke settings were often being changed. In addition, many water injection wells either did not have down-hole pressure gauges, or do not have detailed pressure records to permit more complete analysis of the back-flow period. The back-flow of future water injection wells will generate less revenue than previous back-flow because the new injectors will be in deeper layers and are expected to produce some water almost immediately. Acid stimulation was conducted for the past wells, and this process is considered sound and will be continued. Back-flow is estimated to add some $700,000 to the cost of a water injector because a full production hook-up, including lift-gas, is required. This cost will, however, be more than justified if a 5% increase in injectivity can be shown to result from back-flow.