Today, disease surveillance protocols fail to improve herd health due to a lack of compliance of farmers. An explanatory factor could be that farmers and scientists who design these protocols do not use the same indicators for health surveillance. The aim of this research was to describe how various health indicators differ between farmers and scientists when designing a herd health surveillance tool. Twenty semi-structured research interviews were conducted with French organic dairy farmers and their advisors in animal health. The indicators adopted for herd health surveillance and their characteristics are described and analysed to identify how large the gap is between the indicators used by farmers and those proposed by scientists. Each farmer chose a unique combination of herd health indicators and alert levels. Not one farmer adopted all the indicators as proposed by the scientists. Alternative indicators proposed by farmers and those proposed by farmers can be close to those proposed by scientists or far away, respectively by choosing indicator for earlier detection of disease or indicators measuring very farm specific health management measures for example. Furthermore, analysis of the transcribed interviews allowed identifying a variety of reasons that explain the choice of alternative indicators proposed by farmers. Thus, the results show a great heterogeneity amongst farmers in indicators used for herd health surveillance. Moreover, a gap was identified between what scientists consider as indicators for herd health surveillance and what is adopted in the field. The description of the characteristics of the alternative indicators proposed by farmers and the understanding of some of the reasons for their choices could be used in the design of better herd health surveillance protocols that are more acceptable in field conditions