In accordance with the precept of “information minimization,” many web corporations are opting to document much less information. Nevertheless, that is usually at odds with A/B testing efficacy. For experiments with models with a number of observations, one standard data-minimizing approach is to combination information for every unit. Nevertheless, actual quantile estimation requires the complete observation-level information. On this paper, we develop a technique for approximate Quantile Therapy Impact (QTE) evaluation utilizing histogram aggregation. As well as, we are able to additionally obtain formal privateness ensures utilizing differential privateness.