This paper develops a formal model that allows research, development and evaluation from a functional point of view and attending to security needs, of the different parts in the process of creation of a linguistic steganographic system based on the lexical–semantic substitution of words by their synonyms. The advantages and drawbacks of this type of steganographic technique and the procedure to evaluate its usefulness in a given language are discussed. As a case study, the application of this model is carried out in the Spanish language based on the development of the tool Lexical Substitution in Spanish. This tool allows the testing of the substitution of words with their synonyms in Spanish. A hiding capacity of at least 0.1864 bits/word, resulting in 68.33% of the stegotexts created being invisible to the attacks analysed. This hiding capacity is useful enough to send Internet addresses, GPS coordinates, brief mobility messages, signatures on text, control messages, and so forth. All these uses are in fact very practical within the area of telecommunication networks. From the point of view of corporate network security, stegotexts make possible, for example, the control of a botnet without any knowledge on the part of the network administrators. Copyright © 2013 John Wiley & Sons, Ltd.