Cracking Down on Debt Complaints

What Gets You Debt Relief? Presentation with Visualizations

Inspiration

Our inspiration comes from our shared stress of dealing with customer service experiences. We have all noticed that customer care and service across companies have changed significantly following the implementation of AI into company strategies over the past couple of years. We wanted to measure whether this has had any impact and whether that impact has been positive or negative.

What it does

Our models demonstrates trends in customer complaints from the years of 2021 to 2025, known to be significant years in the industry following COVID-19 and the rising shifts in AI. This proves significant information through visualizations in timely response, sub-products, relief impact, and geographic analysis to cover all avenues of customer complaints in companies.

How we built it

We combined CFPB debt collection complaint data from 2021–2025, cleaned and merged the datasets, and analyzed trends in response times and company actions.

Challenges we ran into

Figuring out how we could work together to portray the data as effectively as possible. This was challenging because most of our dataset consisted of categorical variables rather than numerical variables, aside from identifiers.

Accomplishments that we're proud of

How well we worked as a team, even though we didn’t know each other beforehand. We communicated our goals effectively, and everyone’s skill set contributed to the project. We were also impressed by how much we were able to accomplish in such a short amount of time.

What we learned

We learned about the many uncontrollable factors that go into customer care. Even with a less numerical dataset, we found that there were still so many questions to be answers with the information that was given. We also learned about the many type of debts that contribute to customer complaints, while learning to sort categorical data properly.

What's next for Cracking Down on Debt Complaints

In the future, we could further develop our predictive model to identify complaints that are more likely to receive slow responses or no company action. We could also create a priority score to help companies identify which complaints may need additional attention and track these outcomes over time.