In spite of the success of ESM in describing mental disorders in context, there are limitations. ESM asks more of the subject than a survey instrument, questionnaire or interview. It demands a look into the private world and experience of the individual over an often prolonged period of time. Creating an environment of trust, required for compliance and accurate reporting, is a crucial and practical part of the Experience Sample process that may not be casually passed over. In this Section, the article discussing compliance in elderly depressed patients is a good example of the care that is required in sampling certain groups. Heroin addicts, studied by Kaplan, serve as a good example of research methods tailored to the life style of the population under investigation. Special care must be taken in creating a research alliance and in selecting appropriate questions that speak to and access the experiental world of a specific group, while remaining true to the diagnostic requirements of the mental disorder under study. In addition, careful attention needs to be paid to planning the timing and intensity of the time-sampling procedure. The ability of patients to comply with the signaling schedule and a ‘good fit’ with temporal nature of the phenomena under study such as stress, panic or ‘getting high’, are crucial considerations. Although smaller than expected initially there are specific limits to the use of the ESM. Difficulties have been encountered in sampling elderly subjects with dementia, individuals with severe melancholia and the acutely psychotic.