Purpose Entrustment has mainly been conceptualized as delegating discrete professional tasks. Because residents provide most of their patient care independently, not all resident performance is visible to supervisors; the entrustment process involves more than granting discrete tasks. This study explored how supervisors made entrustment decisions based on residents' performance in a long-term family medicine training program. Method A qualitative nonparticipant observational study was conducted in 2014-2015 at competency-based family medicine residency programs in the Netherlands. Seven supervisor-resident pairs participated. During two days, one researcher observed first-year residents' patient encounters, debriefing sessions, and supervisor-resident educational meetings and interviewed them separately afterwards. Data were collected and analyzed using iterative, phenomenological inductive research methodology. Results The entrustment process developed over three phases. Supervisors based their initial entrustment on prior knowledge about the resident. In the ensuing two weeks, entrustment decisions regarding independent patient care were derived from residents' observed general competencies necessary for a range of health problems (clinical reasoning, decision making, relating to patients); medical knowledge and skills; and supervisors' intuition. Supervisors provided supervision during and after encounters. Once residents performed independently, supervisors kept reevaluating their decisions, informed by residents' overall growth in competencies rather than by adhering to a predefined set of tasks. Conclusions Supervisors in family medicine residency training took a holistic approach to trust, based on general competencies, knowledge, skills, and intuition. Entrustment started before training and developed over time. Building trust is a mutual process between supervisor and resident, requiring a good working relationship.