Annotations can be helpful for analysts in analyzing large datasets and gathering insights. This thesis focuses on developing an annotation system supporting keyboard and touch pen device. The primary purpose of this system is to facilitate annotating visualizations, comparing performance and preference between handwritten annotations made using a touch pen devices and annotations entered via keyboard input. The annotation system includes interactive features that enhance the analyst’s sensory discovery process. By providing annotation capabilities, the system allows analysts to add personal insights, interpretations, and contextual information to visualizations, helping them better understand the underlying data. This research also includes a study to assess performance for handwritten and typed annotations. The study involves participants from various backgrounds and data analysis expertise to perform data analysis tasks using the annotations as the artifacts for the presented system. Using the Think Aloud method, participants verbalize their thoughts, decision-making processes, and perceptions while interacting with the system. Objective performance measures, such as completion time and accuracy, are recorded and compared between the two annotation modalities. I also collect subjective metrics such as preference, satisfaction, and perceived ease of use through post-task surveys and interviews. This qualitative and quantitative data allows us to assess the strengths and weaknesses of keyboard-based and touch pen-based annotation methods. The results of this study contribute to the existing body of knowledge on the effectiveness of different annotation modalities in visualization. The requirements of the annotation system using keyboard and touch pen are collected from the study data. The results also compare the externalization count made using keyboard and touch pen. The results reveal the strengths and limitations of handwritten and typed annotations and provide valuable insights for researchers and practitioners in choosing annotation methods for their projects. The observations and participants feedback reflects that the hardware limitations impact the annotation behavior and experience, and hence need to be addressed in future systems. Ultimately, the annotation system developed in this thesis allows analysts to explore and comment to gain deeper insights, facilitating the evidence-based decision-making process.