Information security is one of the most important methods of protecting the confidentiality and privacy of internet users. The greater the volume of data, the more the need to increase the security methods for protecting data from intruders. This task can be challenging for researchers in terms of managing enormous data and maintaining their safety and effectiveness. Protection of digital content is a major issue in maintaining the privacy and secrecy of data. Toward this end, digital watermarking is based on the concept of information security through the insertion and detection of an embedded watermark in an efficient manner. Recent methodologies in the research on digital watermarking result in data distortion after embedding the watermark. This scenario can reduce the efficiency of detecting watermarks as well as violate data confidentiality. This study adapts a robust reversible histogram shifting (HS) technique for textual data in relational databases. Furthermore, the study presents a digital watermarking scheme intended for document copyright protection and proof of ownership. The major advantage of the proposed watermarking scheme is that it can protect digital data and preserve data quality. To the best of our knowledge, this research methodology is promising for use in the HS expansion model for watermarking data on non-numeric databases. In addition, the results showed that despite exposing the database to an insertion attacks at 50% and 75% of the watermark, the proposed algorithm can recover more than half of the embedded watermark in all addition and alteration attacks cases. As a result, the watermark information can be retained and restored completely.