In this current system of lending, there are more non-bank lenders than ever before. In this new and chaotic space, Team Frugal has developed a natural language processing model for classifying consumer complaints by the nature of the issue to give lenders a tool to find directions for improvement using their complaint databases, while also giving consumers a clearer picture of what kinds of problems plague certain lenders. Using data from the Consumer Financial Protection Bureau that has been manually classified by issue, our project utilizes the RoBERTa model for NLP to build a model that can take consumer complaint data regarding loans and classify it according to issue.