The purpose of this study is to find the variables that explain the behavior of daring to recommend a brand of telecommunications services, from demographics, consumer satisfaction by interactions at various stages of the client cycle, to suggest key points in the business. The development of the study is based on the CRISP methodology and data mining techniques were applied (such as classification trees CHAID). Patterns were found from the answers contained in the survey of recommendation and satisfaction -applied during 2015-, focuses in NPS_General (goal variable), field which classifies customers as Promoters , Passives or Detractors -that shown affinity to recommend a brand-, and other variables as predictors.