Purpose: The purpose of this study was to examine the architecture of internship coaching models from five innovative principal preparation programs in the Southeastern region of the USA. The researchers used coaching architecture in this context to include the assignment of coaches to interns, dosages, and enactment of evaluation and confidentiality. Design/methodology/approach: The researchers used a qualitative, collective case study research design that included semi-structured interviews of program directors and coaches from each of five programs, totaling 19 interviews that lasted from 30 min to an hour and 15 min each. Findings: Commonalities among architectures of principal intern coaching designs included coaching assignment by geography, frequency and format of coaching sessions and length of the internship. All five programs recommend continuing coaching into initial years of administration. Points of distinction pertained to the utilization of external versus internal coaches, confidentiality and evaluation by coaches. Research limitations/implications: This study may inform coaching models for principal preparation programs within similar contexts. Because all five programs are grant-funded within one US state, generalizability and transferability cannot be assumed. Practical implications: The authors provide design considerations for coaching programs, as well as policy considerations and directions for future research. Originality/value: While coaching is increasingly used in leadership preparation programs, there is a paucity of research regarding the nature of coaching models, especially in terms of their architecture. The researchers examine, compare, and contrast coaching model architecture, raising important considerations for coaching designs.