A data set of 30000 records and 24 variables containing information on defaults, demographic factors, credit data, delinquency, repayment and billed amounts of a credit card client in Taiwan from April 2005 to September 2005. The objective was to apply statistical, data visualization and Machine learning techniques (supervised and unsupervised) on the data set to come up with insights and recommendations that could have been used to control bad debts.