Evaluation of fracturing treatments in tight-gas sands is frustrated by uniqueness problems when matching gas rates and bottomhole-flowing pressure (BHFP) data alone. A fit-for-purpose simulator was used to match gas rates, BHFP, cumulative produced water, and composition of the water during five days of cleanup after a fracturing treatment. The simulator models physics of flow and chemical physics to provide improved history matching of the cleanup process. Downhole gauges were in place during cleanup and a 16-hour, shut-in period. History matching of the well returns data with the fracture cleanup model provided multiple solutions when matching gas, water, and BHFP. These solutions all fell on a “line-of-solutions” using a special plot of reservoir properties vs. conductive length. Pressure transient analysis (PTA) of the gas rate and BHP during cleanup and buildup provided solutions that also fell on the same “line-of-solutions.” A “root-of-time” plot of the 16-hour buildup indicated a fracture of infinite acting conductivity and no fracture face-skin damage. A “productive proportion” plot of the simulated fluid recovery vs. the ratio of conductive to created length was used to help narrow, or break, the uniqueness problem in the line-of-solutions. These tools demonstrated that the conductive length was roughly 400 ft. The reservoir permeability to gas was much lower than expected and was the cause of the disappointing productivity. This work demonstrates that matching well-return compositions with a fracture-cleanup model capable of chemical physics can lead to improved interpretations of both reservoir properties and fracture structure. Three new graphical tools provided compelling evidence for the interpretation, and should have significant value in future evaluations.