Recent studies report low and variable phosphorus (P) fertiliser use efficiency (PUE) for cotton in the northern grains region (NGR) of eastern Australia. This may be due to cotton accessing P pools that are not currently tested for in the subsoil (10–30 cm) or variation in response to P source and placement strategy. Two glasshouse studies were used to investigate this, incorporating two soil P tests to assess readily and slowly available P pools (Colwell, and a dilute acid colloquially referred to as the BSES extractant), and five different P fertiliser placement strategies in the subsoil. Eighteen Vertosols were collected across southern to central Queensland in the NGR, and then used to grow faba bean (Vicia faba L.) and cotton (Gossypium hirsutum L.) sequentially in the same 28-L pot. Readily available P pools assessed by Colwell-P were of major importance for faba bean and cotton dry matter, as well as for tissue P concentrations. Cotton was less responsive to extractable subsoil P concentrations than faba bean, suggesting either greater internal PUE or improved ability to accumulate P under conditions of limited availability. We recommend that subsoil P fertilisation should occur before sowing faba bean to maximise PUE in a cotton–faba bean rotation. Faba bean and cotton both recovered more P when the subsoil was fertilised, but no individual P fertiliser placement strategy was superior. Phosphorus extracted using the BSES method was not correlated with faba bean or cotton dry matter or tissue P concentration over the single crop cycle. We also recommend that Colwell-P be measured in the topsoil and subsoil to understand the quantity of plant-available P in Vertosols of the NGR, and that further research is needed to describe the resupply of the readily available P pool from slowly available P pools during a single crop cycle.