In order to enable power companies to grasp the trend of electricity consumption of the whole industry from the power source, this paper proposes a forecasting method of industry electricity sales based on capacity utilization characteristics and taking external economic factors. Firstly, it separates operation capacity of each industry and restores real memory capacity by using the growth curves of different types of business expansion. Then, it extracts industry capacity utilization hours, identifies the external factors which affect the industry utilization characters by correlation coefficient method and K-L information method to determine the dominant factor indexes. At last, it creates power prediction method based on capacity utilization characteristics. The prediction model of industry electricity sales is based on the generation of electricity generated from the stock and the industrial expansion. The results of the case study show that the prediction method can effectively improve the prediction accuracy.