In today’s modern world, mobile has turned out to be one of the essentials for all the people irrespective of their status and profession. With the help of this device, all the data about a person can be tracked down (i.e. from the diet, diseases to contacts and transactions). In case of criminal investigations, inspectors need to collect information about a victim or accused. For this, the individual’s mobile phone plays a vital role. However, it is very difficult to access a device without its owner’s permission. Suppose, if the victim is dead or not willing to expose the information, the cyber police should perform many complex tasks to retrieve the data. Thus, there is a great need to analyze this task and make it feasible to find out pass-codes in order to access the mobile device. This paper explains how passwords can be cracked with ease with the help of survey and training of large dataset. We all know that people these days are very active on social media, which makes it easy to track them. In this work, we analyze a few passcodes and patterns and try to test that data by giving the queries. Here, we can analyze and retrieve the data from an individual’s social media such as date of birth, name, personal information and try to predict the passcode in very few attempts as it turns out that majority of the time, the passcode is generally predictable based on some key characteristics identified in this paper. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group.