Gas cyclic-pressure pulsing is an effective improved-oil-recovery (IOR) method in naturally fractured reservoirs. A limited number of studies concerning this method in the literature focus on specific reservoirs, yet the optimum operating conditions have not been broadly investigated. In this study, we present a detailed parametric study of the process from both operational and reservoir perspectives. Incremental oil production, discounted incremental oil production, and net present value (NPV) are considered as the important markers for the performance criteria. The necessary analyses are performed using a single-well, dual-porosity, compositional reservoir model. In the first part of the study, parametric studies are conducted to develop a better understanding of the operational parameters affecting the process performance in the shallow, naturally fractured, and depleted reservoir of Big Andy field in eastern Kentucky, USA. These include analyses of various design parameters (e.g., soaking period, cycle rate limit, number of cycles, cycle, and cumulative injected-gas volumes). In the second part of the study, reservoir characteristics are investigated. Comparative discussions are presented between cases with CO(2) and N(2) as the injected gas on reservoir fluids of different compositions (heavy, black, and volatile oils). Influences of area, thickness, fracture/matrix permeabilities, initial reservoir pressure, and temperature on the process are studied. It is observed that N(2), as a lower-cost gas, would be a better choice than CO(2) in the Big Andy field. With the oil price used in this study, the cost of injected gas becomes relatively insignificant in economic considerations. Increased income from increased oil production overcomes the increased costs with higher volumes of gas. The way reservoir characteristics affect the process performance is similar in cases with CO(2) and N(2), but differs significantly with different reservoir fluids. Thicknesses ranging between 20 and 50 ft produced more favourable results than thicker reservoirs. A higher efficiency was observed with smaller drainage areas (5 to 8 acres) in the presence of heavy oil. For the cases with volatile and black oil, it is observed that the process efficiency is not altered significantly by the area. The phase behaviour of the reservoir fluid is important for the performance of the process. Initial pressure/temperature of the reservoir and, therefore, the initial fractions of gas/liquid phases affect the process efficiency in a more pronounced manner.