The work aims to analyze the reasons for the difference between the quality level and the published data through the repetitive test of impact resistance items of beer bottles and explore the countermeasures suitable for continuous supervision and reducing the re-inspection. Through testing 99 groups of samples from different regions, manufacturers, and specifications under the same condition, the reproducibility of item results was counted. The result reproducibility of continuous testing of anti-impact items under the same conditions, whether passed or failed, was very low. Similarly, if the spare samples were used for re-inspection in supervision, and the re-inspection result was taken as the final conclusion, it would cause a high probability of negative conclusion, which may increase the difficulty of supervision, and put the inspection institution in an embarrassing context when issuing the report. As a destructive test item with low repeatability, the impact index of beer bottles is not suitable for re-inspection of spare samples. The test process should be strictly standardized and the image records of field test should be kept to be used as the proof data of objection of re-inspection. It should facilitate continuous supervision and avoid self-denial, but also safeguard the rights and interests of all parties and ensure the credibility of supervision.