The first thing to consider is the technical architecture of the project, what kind of interface is used, whether the data transmission of the interface involves security, whether it is necessary to encrypt and decrypt the transmitted data, whether it is necessary to carry out multi-protocol conversion, etc. How to verify the correctness of data (such as the number of RETURN, DATABASE, APPLICATION LOG, etc.) is something that needs to be clearly understood before doing interface testing.

Secondly, in the process of testing, from small to large, from simple to complex, the specific single business (function) interface should be considered first. Every transaction interaction process involves quite a few subsystems or interactions between modules. First, the interface of a single module \ subsystem should be tested to confirm whether the interface is aimed at an external system or an internal subsystem or between modules. In particular, it is necessary to test the interface content with specific business.

Finally, the business process of the whole system is tested, and the process is tested in series, focusing on some abnormal process processing of docking, such as success between subsystems, timeout, and subsequent process processing of corresponding error codes.

In the rapid iteration process of subsequent projects, the use cases of interfaces will be continuously maintained and added according to the demand design and development progress, so as to ensure the stability of interface functions in the later iteration update process of the project. This is a very important point when there are a large number of interface tests in the system (adapting to agile project testing through real-time build-regression testing), and it is also a very important aspect to reflect the value of automated testing. In addition, in order to realize the hierarchical and overall planning of the automation process.