The Layered Communication Model (LCM) describes intersubjective development based on eight different communicative behaviors in three consecutive layers. Earlier studies showed that when the model is used in an intervention, the presence of many LCM behaviors increases from before to after the intervention. The present study aims to relate the content of the intervention to its effectiveness to learn whether and how the LCM can be used to improve communication. 15-minute videos of four student-teacher dyads were coded in ten-second intervals for the presence and absence of the eight LCM behaviors before, during and after the intervention. The intervention was divided into two phases: self-assessment and video feedback coaching. Intervention content was described based on the behaviors that were targeted for improvement during the two phases. Effectiveness was measured by calculating the percentage increase in presence between phases and by calculating effect sizes using a nonoverlap of all pairs method. Results showed that the second intervention phase (video feedback coaching) was most effective in terms of increasing the presence of LCM behaviors and creating larger effect sizes. Effectiveness measures decreased during the follow-up phase but were still higher than at baseline. Furthermore, effectiveness was higher for targeted behaviors than for untargeted behaviors. In conclusion, the LCM can be used as a tool to improve communication, especially when specific behaviors are clearly targeted and video feedback coaching is used to clarify how to work on improving the presence of those behaviors. The self-assessment phase needs adjustments to increase its effectiveness.