Sampling error can cause problems if they are not taken care of. Errors in judgment about sample size can be fixed easily and sample sizes must be considered seriously if big data is being used for predictive analysis. A leader trying to use big data in predictive analysis should always consult the data scientist. The way to understand whether enough data has been collected or not for the purpose of prediction involves understanding the tolerance of the risk associated to accept the assumptions drawn from the sample size characteristics. There are two types of risk: the risk that you're going to take some action when you shouldn't and the risk that you are not going to take some action when you should. Also enough information should be available about the sample variation and precision of measurement to know whether enough data has been collected to make prediction. To know more about importance of sample size in predictive analytics, go to John Weathington (President and CEO of Excellent Management Systems, Inc.)'s link: http://www.techrepublic.com/blog/big-data-analytics/why-samples-sizes-are-key-to-predictive-data-analytics/