Phase I: Analyzing Users, Competitors, and Initial Designs

Introduction

Getting a credit card is an important financial decision for almost every American. However, credit cards can be complex and choosing the right one can be overwhelming. The goal of our website, Credit Card Match, is ultimately to help users become familiar enough with credit cards so that they can confidently choose their next card and not doubt or regret their decision. For this phase of the project, we focused on analyzing our competitors and creating the rough draft for the design of our website.

Methods

The first research method we conducted was competitive analysis. We selected 4 sites and analyzed their strengths, weaknesses, quality level, the platforms they are on, and how much they cost. The sites we chose were NerdWallet, The Points Guy, Bank of America, and WalletHub. We chose these sites because they are all popular and focus heavily on finance and credit cards. NerdWallet in particular has a very detailed portion of its website dedicated to credit cards, so we thought this site would be our closest competitor. All sites were free and worked on web so our analysis mainly focused on each site's strengths and weaknesses. One major weakness we found was that each site made money off their content or recommendations, which opens up opportunities for conflicts of interest. These sites were all also fairly cluttered and not focused specifically on credit cards like our site was, meaning it was more complex than it needed to be for users just interested in credit cards. Through our analysis we realized we could provide something that all these other sites couldn't, which was a minimal and conflict of interest free website designed solely to inform people on credit cards.

Next, we conducted a heuristic evaluation with the following heuristics (n=11): visibility of system status, match between system and the real world, user control and freedom, consistency and standards, error prevention, recognition rather than recall, flexibility and efficiency of use, aesthetic and minimalist design, help users recognize, diagnose, and recover from errors, help and documentation. These heuristics cover most of what our site needs to focus on and good design practices in general. We ended up finding the site creditcards.com, which was much more similar to what we envisioned our site to be than the websites from the competitive analysis. Because of that, we decided to perform the heuristic evaluation on creditcards.com instead of a website from the competitive analysis. We found the site to be quite easy to use, even as new users, and it scored strongly in most categories. Our main areas of critique were that navigating the site could be made easier with shortcuts and buttons, and the site could be made in a more minimal fashion so there is less clutter to potentially confuse users.

Findings

Through our competitive analysis of NerdWallet, The Points Guy, Bank of America, and WalletHub, we found that simple interfaces and organized information were common strengths. Nerdwallet offered a broad card selection, The Points Guy provided a more focused expereince for travel rewards, and Bank of America used customer data to recommend cards. However, NerdWallet included many unrelated financial tools, The Points Guy had a narrow focus, and Bank of America only recommended its own cards. Affiliate relationships could also influence recommendations on sites such as NerdWallet and The Points Guy. These findings suggest that our users need a focused and unbiased service that recommends cards from multiple issuers based on their spending habits and goals.

During our heuristic evaluation of creditcards.com, we found it scored highly in the Match Between the System and Real World, Consistency and Standards, and Flexibility and Efficiency of Use heuristics. The familiar star rating system and catalog style layout made the site easy for users to understand and navigate. It also presented important details such as introductory offers, annual fees, regular APRs, and recommended credit ranges in the same format across cards, while following conventions used by other credit card websites. Its filters and card comparison tool also helped users narrow down their options and compare relevant cards efficiently. However, it scored lower in the Recognition Rather Than Recall heuristic because its tooltips and expandable sections contained large amounts of text that could overwhelm users. All of these findings suggest that our users need a familiar and consistent interface that makes cards easy to filter and compare, while keeping detailed explanations clear and concise.

Conclusions

Credit Card Match should provide a simple questionnaire that gives unbiased, personalized card recommendations from multiple issuers based on users' spending habits and goals. Each card should present its key details in a familiar and consistent format. Additional information about each card's rewards, introductory offers, benefits, and terms within tooltips and expanded sections should be concise and easy to understand. Filters and comparison tools should help users quickly find options that fit their needs. These design recommendations should guide future work toward a focused expereience that gives new users clear guidance while allowing experienced users to compare cards efficiently.

Caveats

Despite our findings, there are several limitations to keep into consideration. Our competitive analysis only covered four websites, which only reflects a subset of the possible credit card comparison services that exist and thus may fail to consider the needs of certain users. Our analysis also focused mostly on visible features, which does not account for how these sites actually recommend cards in the backend nor how accurate the information is. Furthermore, the website we chose to perform our heuristic evaluation on, creditcards.com, was not one of the sites featured in our competitive analysis, so there is limited assessment of its usability compared to other services we analyzed.

Secondly, because we only performed a heuristic evaluation, we have limited knowledge regarding what features are actually effective for real users. While we have a good baseline for establishing a minimum viable product, it will require feedback from users to truly identify what our target audience prefers or is frustrated by. Overall, our conclusions have led us to some strong design prototypes, but further testing and research will be required to solidify a recommendation algorithm, easy-to-use interface, and the kinds of resources to have accessible for users to feel confident in their decisions.