Background Although for more than a decade healthcare systems have attempted to provide evidence-based mental health treatments, the availability and use of psychotherapies remains low. A significant need exists to identify simple but effective implementation strategies to adopt complex practices within complex systems of care. Emerging evidence suggests that facilitation may be an effective integrative implementation strategy for adoption of complex practices. The current pilot examined the use of external facilitation for adoption of cognitive behavioral therapy (CBT) in 20 Department of Veteran Affairs (VA) clinics. Methods The 20 clinics were paired on facility characteristics, and 23 clinicians from these were trained in CBT. A clinic in each pair was randomly selected to receive external facilitation. Quantitative methods were used to examine the extent of CBT implementation in 10 clinics that received external facilitation compared with 10 clinics that did not, and to better understand the relationship between individual providers' characteristics and attitudes and their CBT use. Costs of external facilitation were assessed by tracking the time spent by the facilitator and therapists in activities related to implementing CBT. Qualitative methods were used to explore contextual and other factors thought to influence implementation. Results Examination of change scores showed that facilitated therapists averaged an increase of 19% [95% CI: (2, 36)] in self-reported CBT use from baseline, while control therapists averaged a 4% [95% CI: (-14, 21)] increase. Therapists in the facilitated condition who were not providing CBT at baseline showed the greatest increase (35%) compared to a control therapist who was not providing CBT at baseline (10%) or to therapists in either condition who were providing CBT at baseline (average 3%). Increased CBT use was unrelated to prior CBT training. Barriers to CBT implementation were therapists' lack of control over their clinic schedule and poor communication with clinical leaders. Conclusions These findings suggest that facilitation may help clinicians make complex practice changes such as implementing an evidence-based psychotherapy. Furthermore, the substantial increase in CBT usage among the facilitation group was achieved at a modest cost.