In recent decades, a growing interest in learning Spanish as a foreign language has grown. The impact of socioeconomic and sociocultural events on Spanish society (internationalization of the Spanish economy, opening of Cervantes Institute, Erasmus students, linguistic tourism, and an increased interest in Spanish culture) has led to new lines of research derived from the teaching of Spanish as foreign language. The article we present tries to serve as a guide for teachers to be able to diagnose, classify and solve the characteristic problems of learners of Spanish. The phonetic, morphological, lexical, grammatical and pragmatic peculiarities of their mother tongue interfere and hinder the process of learning Spanish. The research is oriented in its different stages to collect, analyze and provide results from written compositions in students’ blogs. Specifically, the research aims at the design of a methodology based on the principles of corpus linguistics in its different stages, from the corpus compilation, its labeling (identification and classification of mistakes) and note taking until the extraction of the results. The main result is the creation of a taxonomy and its labeling subdivided into four major subsections: lexical, grammatical, discursive and graphical errors.In recent decades, a growing interest in learning Spanish as a foreign language has grown. The impact of socioeconomic and sociocultural events on Spanish society (internationalization of the Spanish economy, opening of Cervantes Institute, Erasmus students, linguistic tourism, and an increased interest in Spanish culture) has led to new lines of research derived from the teaching of Spanish as foreign language. The article we present tries to serve as a guide for teachers to be able to diagnose, classify and solve the characteristic problems of learners of Spanish. The phonetic, morphological, lexical, grammatical and pragmatic peculiarities of their mother tongue interfere and hinder the process of learning Spanish. The research is oriented in its different stages to collect, analyze and provide results from written compositions in students’ blogs. Specifically, the research aims at the design of a methodology based on the principles of corpus linguistics in its different stages, from the corpus compilation, its labeling (identification and classification of mistakes) and note taking until the extraction of the results. The main result is the creation of a taxonomy and its labeling subdivided into four major subsections: lexical, grammatical, discursive and graphical errors.