This article explores how P (programmed learning) in Revans' formula L=P + Q accumulates from one action learning research initiative to another. Drawing on three inter-organizational action learning research initiatives, it shows how the L (learning) from conducting action learning in an initiative in one network built new P on network action learning research which was applied in two subsequent initiatives. The article contributes an understanding of how P accumulates from learning initiative to learning initiative and how its application contributes to the L of actionable knowledge.