China is well known for providing official data, but how to treat these data is a longstanding debate among China scholars. This paper advances understandings of how to interpret Chinese official statistics about the internet. Using standards for evaluating surveys in the social sciences, we systematically compare official data from the China Network Information Center (CNNIC), which is under the supervision of China’s main regulator of internet policy, with the China Internet Survey 2018 (CIS), which is, to our knowledge, the first nationally representative survey on internet use in China. Using three examples, we illustrate how methodological differences in sampling design and measurement can lead to vastly different conclusions about key indicators of internet use in mainland China, including the percentage of internet users, their regional and urban–rural digital divide, and the percentage of specific social media platforms. We discuss the challenges of survey work on internet use in China and offer recommendations on how to interpret official statistics, especially in light of the limitations researchers face when conducting face-to-face surveys in China.