Precision soil management is fundamental to the sustainable production of high-quality tea, yet the spatial integration of fertility and heavy metal safety remains a significant challenge. This study aimed to delineate multi-dimensional management zones (MZs) in the tea plantations of Tianmuhu, Jiangsu Province, by evaluating three clustering algorithms: K-means (KM), Fuzzy C-means (FCM), and Iterative Self-Organizing Data Analysis Technique (ISODATA). A total of 70 representative soil samples were analyzed for 10 properties. Descriptive statistics revealed pronounced spatial heterogeneity, particularly for Hg (CV = 71.04%) and P (CV = 61.83%). Pearson correlation and Principal Component Analysis (PCA) demonstrated strong synergistic relationships among organic matter (OM), nitrogen (N), and potassium (K) (r = 0.49–0.69, p < 0.01), which formed a distinct Fertility Factor on PC1. Conversely, PCA identified divergent sources for heavy metals, with Cr primarily governed by pedogenic processes (PC2), while Cd were associated with anthropogenic inputs. Guided by these distinct spatial drivers, this study separately delineated fertility and heavy metal safety MZs. The optimal number of clusters was determined by balancing statistical validity with spatial operationality via the Silhouette Coefficient (SC) and Smoothness Index (SI), with results indicating that a 2–3 zone scheme yielded the most favorable scores. Comparative analysis showed that for soil fertility, ISODATA outperformed KM and FCM by effectively capturing the high variability of P and producing statistically distinct zones (p < 0.05). For heavy metal pollution, FCM provided better partitioning by reflecting the continuous gradients of composite contaminants. Validation results showed that while 61% of the area was classified as high-fertility (ISODATA), approximately 63–75% fell into relatively higher heavy metal accumulation categories. This dual-objective zoning framework provides a scientific basis for site-specific fertilization and targeted environmental monitoring in the regional tea industry.