Central bank settlement systems are vital for payment intermediation and have an important role for monetary policy and financial stability. Payments settling in these systems stem from foreign exchange (FX), securities and interbank trading, tax collection, government social support, and purchases of goods and services. A good understanding of the kinds of payments that settle in a central bank real time gross settlement (RTGS) system is useful for both overseers and operators, but no study exists that attempts to systematically categorize all payments settling in an RTGS system. In this paper, we have categorized payments that settled in Norges Bank’s settlement system (NBO) during one month, according to the payments’ origins. Our work was enabled by the collection of transaction-level data from various financial market infrastructures (FMIs) and NBO participants, data mapping and the use of algorithms. 80% (77%) of all payments by number (value) were categorized confidently. Allowing for some uncertainty and lower precision, inference of the remaining 20% (23%) was made, leaving only 10% (8%) of payments uncategorized by number (value). Interbank lending, FX trades, and payments involving households and companies were the major contributors to liquidity use in NBO.