Wearable devices offer unique opportunities for self-management of chronic conditions. However, long-term wearable data is often underutilized due to its complexity. In this work, we investigate opportunities around long-term wearable data from continuous glucose monitors (CGMs) used for diabetes management. Our study includes analysis of up to 7 years of wearable CGM data from 11 people with type 1 diabetes (T1D), followed by semi-structured interviews to unpack factors that affect management. Using seasonal visualizations as probes during the interviews, we found that blood glucose management is highly individual with the same factors yielding divergent outcomes across participants. We also found that seasonal visualizations of an individual’s long-term data across various time horizons (e.g., months of the year, days of the week, hours of the day) can improve awareness of trends, prompt reflections on influencing factors, and inspire actionable strategies to improve outcomes. Insights from this study highlight untapped opportunities around leveraging long-term wearable data in diabetes management and inform design implications for personal informatic solutions across other continuous monitoring domains.