In concentrated solar power plants using direct steam generation, the usage of a thermal storage unit based only on sensible heat may lead to large exergetic losses during charging and discharging, due to a poor matching of the temperature profiles. By the use of the Kalina cycle, in which evaporation and condensation takes place over a temperature range, the efficiency of the heat exchange processes can be improved, possibly resulting also in improved overall performance of the system. This paper is aimed at evaluating the prospect of using the Kalina cycle for concentrated solar power plants with direct steam generation. The following two scenarios were addressed using energy and exergy analysis: generating power using heat from only the receiver and using only stored heat. For each of these scenarios comparisons were made for mixture concentrations ranging from 0.1 mole fraction of ammonia to 0.9, and compared to the conventional Rankine cycle. This comparison was then also carried out for various turbine inlet pressures (100 bar to critical pressures). The results suggest that there would be no benefit from using a Kalina cycle instead of a Rankine cycle when generating power from heat taken directly from the solar receiver. Compared to a baseline Rankine cycle, the efficiency of the Kalina cycle was about around 5% lower for this scenario. When using heat from the storage unit, however, the Kalina cycle achieved efficiencies up to 20% higher than what was achieved using the Rankine cycle. Overall, when based on an average assumed 18 hours cycle, consisting of 12 hours using heat from the receiver and 6 hours using heat from the storage, the Kalina cycle and Rankine cycle achieved almost equal efficiencies. A Kalina cycle operating with an ammonia mole fraction of about 0.7 returned an averaged efficiency of about 30.7% compared to 30.3% for the Rankine cycle.