Procrastinate is to voluntarily delay an intended course of action despite expecting to be worse off for the delay. Despite progress in delineating characteristics and correlates of individual differences in procrastination tendencies, relatively neglected questions concern the decision of procrastination. As any attempt to procrastinate things inevitably involves the decision between doing it now or doing it later, this kind of decision can be considered as a key component of predicting procrastination. Thus approaches about predicting procrastination can be further simplified as a decision making process of determining doing things now or doing things later. In order to disclose the cognitive mechanism of deciding “do it now or later”, we proposed a decision model for procrastination in current manuscript. In this model we argued that: (1) Motivational competition between procrastination and timely engagement is the rooted process for determining to prolong a task or not. Specifically, humans’ s tendency to procrastinate should mainly stem from the task averseness when engaging the specific task. On the other hand, a motivation to timely undertake a task usually stems from the related rewards the task can bring or associated punishments which ones are trying to avoid through completing this task. (2) Therefore, this motivational combat can be simplified as comparison between averseness of engaging in task and rewards (or punishments) from the task. In fact, the things we procrastinated usually consist of a aversive process of engaging in them and meaningful results which bring us rewards related to task completion or enable us to avoid punishments associated with task failure. Both pursuing reward and avoiding punishments (Utility of Outcome) impel humans to complete tasks as soon as possible. On the contrary, the aversive process of engaging (Utility of Engaging) hinders humans from undertaking tasks timely. Therefore, to procrastinate or not can be predicted by the comparison between Utility of Outcome and Utility of Engaging from tasks. (3) last but not least, putting off an annoy task to have it’s utility discounted is the reason for procrastination. It is humans’ propensity to overvalue instant rewards and discount the outcomes of distant future. Traditional perspectives suggested extremely discounting outcomes (reward or punishment) of a task in the far future lead to procrastinating this task. However, this view is incomplete because delay discounting not only works for far outcomes, but also has effects on Utility of Engaging when putting off a task. The discounting of far outcome of a task is an unchangeable feature rooted in task itself. However, voluntarily putting things off or not makes it quite controllable whether to have task averseness stemming from timely engaging discounted or not. Finally, in order to make our points more clear, examples and application of decision model for procrastination were discussed. Generally speaking decision model for procrastination is able to further the understanding of procrastination and be helpful to untangle the cognitive mechanism of procrastination and modulations of influence factors respectively. Implications of decision model for procrastination should be extensive, we speculated a comprehensive framework of procrastination and it’s influence factor can not be far.