Citizen science programs in ecology have greatly developed in France since the 2000s. However, the diversity of skill levels of participants in such programs may prevent the design of complex protocols. The FEDER program for the conservation of the Reunion Harrier on Reunion Island aimed at answering several scientific questions. Some of them required the use of advanced methodologies, notably for estimating and correcting detection bias. We designed a protocol in several stages intended for volunteers, professionals and students with varied ornithological skills. Among the 150 people who initially enrolled in this program, 103 collected data in May and June 2017 on 330 “survey points” distributed throughout the island. Prior to the field survey they had received training in the classroom and in the field in order to reduce variation in their ornithological skills and to ensure that everyone understood the objectives and the data collection protocol. At the end of the field surveys, a questionnaire completed by 71 participants allowed us to collect information about their motivations and the difficulties they faced. In this article we report the different steps required for the design and the deployment of this program and we analyse its strengths and weaknesses, in terms of establishment and retention of a participant network. This analysis is for the use of other citizen science programs designers. We emphasize the importance of project officers actively sustaining participant engagement for ensuring the success of such a program.