Aims Tacrolimus is a critical dose drug and to avoid under‐ and overexposure, therapeutic drug monitoring is standard practice. However, rejection and drug‐related toxicity occur despite whole‐blood tacrolimus pre‐dose concentrations (Tacblood) being on target. Monitoring tacrolimus concentrations at the target site (within peripheral blood mononuclear cells; Taccells) may better correlate with drug‐efficacy. The aim of this study was to (1) investigate the relationship between Tacblood and Taccells, (2) identify factors affecting the tacrolimus distribution in cells and whole‐blood, and (3) study the relationship between Taccells and clinical outcomes after kidney transplantation. Methods A total of 175 renal transplant recipients were prospectively followed. Tacblood and Taccells were determined at Months 3, 6 and 12 post‐transplantation. Patients were genotyped for ABCB1 1199G>A and 3435C>T, CYP3A4 15389C>T, and CYP3A5 6986G>A. Data on rejection and tacrolimus‐related nephrotoxicity and post‐transplant diabetes mellitus were collected. Results Correlations between Tacblood and Taccells were moderate to poor (Spearman's r = 0.31; r = 0.41; r = 0.61 at Months 3, 6 and 12, respectively). The Taccells/Tacblood ratio was stable over time in most patients (median intra‐patient variability 39.0%; range 3.5%–173.2%). Age, albumin and haematocrit correlated with the Taccells/Tacblood ratio. CYP3A5 and CYP3A4 genotype combined affected both dose‐corrected Tacblood and Taccells. ABCB1 was not significantly related to tacrolimus distribution. Neither Tacblood nor Taccells correlated with clinical outcomes. Conclusions The correlation between Tacblood and Taccells is poor. Age, albumin and haematocrit correlate with the Taccells/Tacblood ratio, whereas genetic variation in ABCB1, CYP3A4 and CYP3A5 do not. Neither Tacblood nor Taccells correlated with clinical outcomes.