In Ghana, the high-risk obstetric referral system is inadequate. Delay is common and patients often arrive to receiving hospitals in compromised states. An effective referral system should include an adequately resourced referral hospital, communication across sectors, accountability, transport, monitoring capability and policy support, which are currently lacking. A pilot program was undertaken to facilitate communication between hospital staffs. Additionally, data was collected to better understand and characterize obstetric referrals in Accra. Thirteen institutions were selected based on referral volume to implement the use of pre-referral treatment guidelines and WhatsApp as a mobile technology communication platform (Platform). Participants included healthcare workers from 8 health centers, 4 district hospitals, the Greater Accra Regional Hospital (GARH), administrators, doctors from other tertiary hospitals in Accra and medical consultants abroad. Facilities were provided smartphones and guidelines on using WhatsApp for advice on patient care or referral. Data were collected on WhatsApp communications among participants (March-August 2017). During this period, 618 cases were posted on the Platform and users increased from 69 to 81. The median response time was 17 min, a receiving hospital was identified 511 (82.7%) times and pre-referral treatment was initiated in 341 (55.2%). Subsequently, data collected on 597 referrals to GARH (September-November 2017) included 319 (53.4%) from Platform and 278 (46.6%) from non-Platform hospitals. Of these, 515 (86.3%) were urgent referrals; the median (interquartile range) referral to arrival time was 293 (111–1887) minutes without variation by facility grouping. Taxis were utilized for transportation in 80.2%; however, referral time shortened when patients arrived by ambulance and with a midwife. Only 23.5% of urgent referrals arrived within two hours. This project demonstrates that WhatsApp can be used as a communication tool for high-risk obstetric referrals and highlights the need to continue to improve urban referral processes due to identified delays which may contribute to poor outcomes.