This study will use second-hand data research for quantitative analysis. These data are obtained from public channels such as relevant Chinese government departments or non-profit organizations such as the China Manufacturing Association, so as to ensure that the information is sufficient to represent most of China’s manufacturing small and medium enterprises, and the data does not contain obvious deviations. Secondary data is data that is usually collected and organized for other purposes. Data collectors are organizations or individuals unrelated to this research. Therefore, the collection of secondary data does not require direct contact between the researcher and the respondent (Vartanian, 2010). Researchers only need to obtain data through public channels. There are many advantages to using second-hand (Jones, 2010). First, the resources of second-hand data are abundant. Generally speaking, widely existing second-hand data can provide cross-year and cross-season longitudinal data, which is conducive to the construction of panel data. Second, second-hand data has the characteristics of objectivity, reproducibility and repeatability. It pursues reliability and scientificity. At the same time, the data itself is heterogeneous, and following academic theories can prevent academic fraud. At the same time, second-hand data are generally from multiple sources, and triangulation is used in the survey process. In addition to avoiding common method deviations, this method can also identify things from multiple angles, thereby more accurately observing and discovering laws (Flick, 2004). Furthermore, using second-hand data for research can collect large amounts of data at a lower cost. In many cases, the data is even free, but researchers need to filter and process the data (Hox & Boeije, 2005). Finally, this study is non-interfering study for previous studies using these data, so there is no impact between the two. There are many sources of secondary data, such as government statistical yearbooks and survey data, international and regional data from international organizations, data from business consulting companies, and data shared by other researchers (Vartanian, 2010). This study uses existing data as research resources to analyze, summarize and rank the obstacles in the supply chain management of my country’s manufacturing SMEs, and propose corresponding improvement measures on this basis. This research is an extensive research and requires a lot of data to support the research, and the use of raw data needs to be independently obtained by the researcher. Therefore, this data collection method is not suitable for this study, as it will consume time and effort. Second-hand data greatly reduces the cost of data collection, and the data collection time is also shorter. At the same time, compared with first-hand data, second-hand data provides a large amount of longitudinal data for this study, which can eliminate the influence of certain specific factors. These factors may be natural disasters, sudden policy adjustments, etc., making the research results more objective.