In modern business process sometime sensitive data is distributed to set of supposedly trusted agents (third parties) by the owner of data. If in case distributed data where found at unauthorized place (e.g. on the website or somebody’s laptop) then the owner must be able to assess the likelihood that the leaked data come from one or more agents or not. Organizations mainly apply data or information security only in terms of network protection from intruders and hackers, but due to globalization and digitization there is rapid growth in the amount of sensitive data processing applications. In the organizations these data can be accessed from different medium which increases the chances of data leakage and trouble in guilt assessment. Data allocation strategy includes fake records in the distribution set along with sensitive data, it acts as watermark to identify the corresponding owner. If such sensitive information is leaked then with the help of fake records and distribution logic involvement of agent in the leakage can be traced. Suitable data allocation strategy improve's the probability of guilt detection and this reduces the catastrophic effect of data leakage.