Pega Crédito

Building a product from scratch

I joined the Pega Crédito team as the sole designer at the very beginning of the project. Although the company was already operating without a digital presence in the lending world, the CEO was achieving surprising results thanks to his deep understanding of the target audience. The rate of people paying their debts was well above the market average. Our primary objective is to obtain data from real users to train our machine-learning model and scale the business. To get there, we decided to build an app from scratch.

🚀

Impacts

Product Launch • +100k Downloads • 70k Loan applications • 400 clients

The challenge

The main challenge was to create an app for a startup that was still in its early stages and needed a well-established visual identity. Additionally, we had a tight deadline of six months to complete the project. While the primary focus was on developing the application, it was crucial to have a logo and visual communication that resonated with the target audience to move forward.

Goals

Obtain user data to train a machine-learning model and scale the business.

🖌️

Develop an identity to communicate with target audiences and guide app design

📱

Build an app to obtain user data and scale the business

🗺️

Mapping the entire experience in a fast-paced environment

The solution

We began developing a visual identity to guide our communication and app design. As we had a tight deadline of three months to launch the first version of the application, we decided to break the project into three parts, prioritized based on the flow of obtaining and paying credit. This approach allowed us to launch the app quickly to gather the necessary data for the machine-learning model while also gradually improving the app over time.

Develop an identity to communicate with target audiences and guide app design

Building the base with the discovery phase

At the start of app development, we aimed for efficiency by gathering relevant information for visual communication and application development.

In collaboration with the project manager, we conducted a structured interview with the CEO and analyzed the communication strategies as well as the application flows of key competitors. These activities provided valuable insights into users, market trends, and the business. Consequently, we could create an effective visual communication strategy and develop the app accordingly.

The system

After this survey, we had enough information to design visual explorations, test communication with the target audience, and then define the brand and visual communication. With this base defined, we could explore the app's screens and define an initial Design System.

Foundation and components Pega Credito Design System

Build an app to scale the business

Onboarding for the Pega Credito Application

Reviewing our knowledge

The app worked on two essential fronts: providing a safe and pleasant experience for those seeking credit and collecting valuable behavioral data to feed our machine-learning model. With these two purposes in mind, we set out to revisit the user insights and competitive analysis we conducted to design the credit application process.

An easy and secure credit application flow

Our main challenge was to offer an experience that conveyed security and was pleasant for those who needed to apply for credit. We knew that people might be skeptical about sharing sensitive information on an unknown app. Furthermore, the credit request flow was extensive due to the diverse information needed to feed the ML model and for compliance reasons.

Mapping the entire experience in a fast-paced environment

The app worked on two essential fronts: providing a safe and pleasant experience for those seeking credit and collecting valuable behavioral data to feed our machine-learning model. With these two purposes in mind, we set out to revisit the user insights and competitive analysis we conducted to design the credit application process.

Changing direction fast

We managed to deliver a safe and easy experience, but the number of requests was growing so fast that our team couldn't keep up. For every 207 requests, we were only able to handle 13 per day, causing delays in processing.

At that time, we were still training our machine learning, and specialized people carried out a meticulous credit analysis. We then stopped the app's development and focused on reducing the frustration of people who continued to request credit and improving our internal credit release tool to speed up this stage.

Conclusion and Results

Throughout our journey, we received over 70,000 loan requests from 100,000 downloads and successfully credited around 400 customers. Unfortunately, we had to close the project due to various reasons, including a lack of resources needed to achieve our business goals.

Skills

Resources

Illustrations by Storyset
Icons by Streamline Icons