Interpreting and creating data visualizations are essential skills in the current data-intensive age. Visualizations are a tool to efficiently extract and communicate information from data. However, novices often have difficulties with visual mapping activities. Understanding these difficulties and investigating ways to improve data visualization design aim to empower novices to explore their data better. In this context, this research aimed to identify and understand gaps in the visualization design process, providing support to novices to select and plan their visualizations more appropriately. With this objective, we conducted two empirical studies. First, we performed an observational study in a visualization design process with five practitioners from different companies. In this study, we mapped the steps of the process, and the material created was evaluated by two visualization experts. Based on this analysis, we identified difficulties and decisions during the process, mainly in choosing appropriate visualizations and mapping variables onto these visualizations. Based on the results and works extracted from the literature, we proposed and evaluated a structured material, VIS-SeMa. The material is composed of topics and uses generic question templates and concrete examples of questions and visualizations to support novices in selecting a visualization type and mapping variables. In the second study, we evaluated the feasibility of the proposed material. The results of the evaluation showed that the material supported most participants in selecting visualizations and creating visual mappings correctly. From this research, we proposed a study process based on analytical questions to evaluate created visualizations and identify decisions, and difficulties faced by novices. Additionally, we proposed VIS-SeMa, a material to support novices in visualization design. The results of the feasibility study showed that participants preferred visual content and that presenting concrete examples helped them understand how to map variables to visualizations. In addition, we observed that combining the anatomy of the questions and visualizations with the examples improved novices understanding.