Time Slotted Channel Hopping (TSCH) enables highly reliable and ultra-low power wireless networking, and is at the heart of multiple industrial standards. It has become the de facto standard for industrial low-power wireless solutions, and a true enabler for the Industrial Internet of Things. In a TSCH network, all nodes remain tightly synchronized by periodically communicating with one another to compensate for clock drift. The synchronization algorithm used in a network determines how often the nodes need to re-synchronize, which greatly influences their energy consumption. This article presents an adaptive synchronization technique which allows a node to learn and predict how its clock is drifting relative to its neighbors’, and coordinates the instants at which the nodes re-synchronize. This technique increases synchronization accuracy, while reducing synchronization communication overhead, thereby extending the battery lifetime of the network. Through simulation, we show how adaptive synchronization allows the nodes in a 3-hop deep network to maintain synchronization within 76μs of one another, while sending an average of only 18.9 re-synchronization packets per hour, a 83% reduction compared to a network not using adaptive synchronization. Through experimentation on a range of hardware platforms, we show how adaptive synchronization is needed for interoperability.