Decoding BofA's Complaint Trends
Understanding the basics first...
CFPB (Consumer Financial Protection Bureau) is a governing body of US government focusing on protecting consumer rights related to financial products and services (mortgage, credit cards etc.). Whenever consumers face challenges related to their financial product/service, they can submit complaints to the CFPB via channels including web, mail, referral, or phone. These complaints are first analyzed by CFPB itself, and then forwarded to the relevant financial institution for resolution.
​
In this case, the relevant financial institution is BofA (Bank of America), one of the largest banking institutions in USA.
​
The data under consideration contains complaints first recieved by CFPB and then forwarded to BofA for resolution. Data spans from 2017-2023, with 62,516 complaints present in the data. The complaints are drilled down in the following order:
1. Product.
2. Sub-product.
3. Issue.
4. Sub-issue.
My job.
I was recently assigned the project, and as a business analyst, my job was to analyze the data and answer the following business questions:
​
1. Do consumer complaints show any seasonal patterns?
2. Which products present the most complaints?
3. Which states have the highest rate of complaints registered?
4. How are complaints typically resolved?
​
I created the dashboard by keeping the business requirements considered - each visual should be focused on answering one of the three questions mentioned above to take better business decisions.
Were there any seasonal patterns?
To answer this question, I created a time-series spline chart to catch any seasonal trend in the complaints registered.
I also created a slider, synced with year slicer to focus on a particular time period. Coupled with that is a brief description for that particular year to give a quick gist to readers on what happened in that year.
​
On a yearly basis, an increasing trend can be observed in the complaints registered. But on a monthly level, I couldn't establish a seasonal pattern as there wasn't any. Safe to say that fluctuations in the variable follow a random variation.
The most complained 'product'.
The page 'Eagle-eye View' gives a holistic view of progress in each country, and contains bar charts in the lower half that communicate the following metrics:
1. Responsiveness in resolving complaints.
2. Registration of complaints:
a. State-wise.
b. Medium-wise.
c. Product-wise.
Product-wise ranking of complaints shows that 'Checking/Savings Account' has been the most complained product, with around >30% of complaints registered only for this product every year. This has been a constant trend for all years, with the mentioned product being the most complained one, followed by the following products ranking at 2nd and 3rd rank from 2017-2023 in no order.
1. Mortgage.
2. Credit/Prepaid Cards.
3. Credit reporting, credit repair services, or other personal consumer reports.
Page 2 of the dashboard, 'State-wise View' gives us a much clear interpretation of the problem through a decomposition tree - consumers have been mostly complaining about their account management, particularly deposits and withdrawals. To understand this, consumers have been complaining the most about facing issues with depositing and withdrawing funds. This can be either:
- Inability to access funds.
- BofA placing holds on deposits.
- Discrepancies between when BofA told consumers their funds would be released and when the funds were actually made available.
A snippet of 2017 is attached; however, similar trend can be observed in all years.
The most complaining 'state'.
Over the years, California (CA) has been receiving the highest number of complaints, and it has been the same trend during 2017-2023. Florida (FL) has been the 2nd highest complaining state, while Texas (TX) and New York (NY) have been at the 3rd rank at different times in 2017-2023.
It is worth mentioning that a correlation can be developed between the population of these states and the quantum of complaints recieved from them: state with higher population has recieved higher number of complaints. Though, in business context, it would make more sense had the correlation been developed between the number of customers in each state and complaints recieved from them. Since this data point was unavailable in the dataset, it is also not included in the dashboard.
The second page of dashboard, 'State-wise View' contains a filled map which provides a comprehensible comparison of complaints in all states at a glance. What is worth noticing is there is no noticeable pattern in number of complaints in states, but rather a non-periodic pattern, meaning that the complaints of Checking/Savings Account have been highest in all the states too.
Complaint Resolution.
Throughout the duration of 2017-2023, BofA has been able to close ~75% complaints with an explanation, with around ~17% closure with a non-monetary relief and 4% closure with monetary relief. It is worth mentioning that the quantum of complaints with a monetary closure has significantly reduced over the years; though it will be worth the effort to observe the impact in absolute figures, which was unavailable in the dataset. The decomposition of resolution product-wise is illustrated in the third page of the dashboard, named 'Resolution'.
