The article analyzes the influence of data source features on the number of sequetial parttern, which includes the number of customers, average transctions of sequences, average items of transctions, and number of different items. The model of single features influence is obtained. With the model, a primary jugdement can be got before mining.