However, by applying SERVQUAL scale to different industries for service quality evaluation, scholars found that there are some conceptual and metric problems in the scale. Ladhar0 and buttli think that consumers' limited cognitive ability makes it difficult for them to form clear service expectations before accepting services, and consumers' service expectations will change dynamically in the process of accepting services. Therefore, it is difficult to practice in Enterprises Accurate service expectation data is obtained, and the gap model of expectation and perception proposed by pa-rasuraman is used to measure service quality. In addition, when some scholars applied the se-rvqual scale to B2C retail, logistics, international marketing channels and other fields, they found that the validity of the scale was not high. It was suggested that the dimensions, indicators and measurement methods of the model should be revised according to the specific characteristics of different industries and enterprises in practical application. In view of this, Cronin and others proposed to delete the expected service quality in the se-rvqual scale, and only measure the service quality according to the perceived service performance, forming the se-rvperf scale. The results show that SERVPERF has higher reliability and validity than SERVQUAL. It should be pointed out that although the se-rvpere scale is also widely used in retail, education, library and other service fields, compared with the generally accepted view that "service quality is the difference between customer's expected service quality and perceived service performance" in the academic community, it is somewhat one-sided to evaluate service quality only through perceived service performance. It can be seen that the superiority of se-rvqual and SERVPERF remains to be further explored.