/ Assumptions.
Built-in functionality in the cash and consolidation loan process used to verify the financial situation of customers applying for credit products through electronic channels
The main objective of the Manual Decision project was to increase the sales volume of cash and consolidation loans through electronic channels.
One opportunity was to offer loans to customers who had previously received a negative credit decision for the amount they requested. Based on analytical data, this group could include several hundred customers per month.
The main problem was insufficient creditworthiness. Some customers wanted to apply for an amount significantly higher than the amount allowed by the bank’s automated credit assessment.
The solution introduced additional mechanisms for analysing the customer’s financial situation. This increased the chance of issuing a positive credit decision for customers who would otherwise not qualify for the requested amount.
The main target group consisted of customers whose credit needs were higher than the amount supported by their current financial situation.
/ Business Impact
The functionality was deployed across all electronic channels where cash and consolidation loans were sold.
The implementation increased the value of sold loans by 12% and increased the number of positive credit decisions by 11%.
The time required for a bank employee to analyse a customer’s financial situation was reduced by 1–2 working days.
The functionality is connected with the PifPaf application, which reduces the time needed to make a credit decision and increases the number of positive decisions during manual analysis of the customer’s financial situation.
/ LOAN VERIFICATION
The process of designing a functionality
/ Discovery
The project started with meetings with the business team to discuss the idea, assumptions and initial requirements.
During design workshops, we defined business and design goals, problems, needs and requirements. We used Business Model Canvas and Value Proposition Canvas to structure the information and better understand the target customers.
Based on existing bank customer segments, we also created detailed personas. To better understand the size of different customer groups, we analysed internal data and compared customers with one negative decision, multiple negative decisions and those with no creditworthiness.
We also conducted an internal competitive analysis of cash and consolidation loan processes to identify similar solutions and useful market patterns.
/ Define
Together with the Service Design Lead, we mapped the existing loan application process using a Service Blueprint. This allowed us to understand both the customer-facing experience and the supporting systems behind the service.
We then created a second blueprint showing how the new solution would be integrated into the existing process.
During this stage, we identified the need for an application used to verify customer creditworthiness and issue credit decisions. The design of this application was treated as a separate part of the project.
To validate the concept and better understand customer experiences after negative credit decisions, I designed an initial prototype of the new functionality and planned qualitative research.
We conducted in-depth interviews with three groups of bank customers: customers who had received a negative decision, customers with no creditworthiness, and a control group of customers who had received a positive decision.
I prepared research scenarios and recruitment requirements for each group. With support from research agencies, participants were recruited and interviewed.
During the interviews, I presented possible versions of the loan application process using prototypes and Miro boards. After the research, I analysed recordings and notes, prepared a report with insights and recommendations, and presented the results to the product team and UX Guild.
The findings were then used in another design workshop, where we updated our assumptions and created the final Service Blueprint. It described the complete service across customer interactions, supporting systems and changes required in existing applications.
/ Develop
System analysts prepared functional descriptions and user stories, in which I actively participated. I also created user flows describing the loan application process with the Manual Decision functionality.
Based on these flows, I designed the final high-fidelity interfaces for mobile and desktop.
I started with the mobile version following the “mobile first” principle and the bank’s existing design standards. At that time, different electronic channels used separate design systems, so I worked with components from three different systems.
The new functionality had to be embedded into the existing loan application process rather than designed as a completely separate experience. Competitive research and market benchmarks supported the design decisions.
A significant part of the work involved designing communication content. Plain language specialists supported the process to make the messages understandable for an average bank customer. The content covered both messages displayed during the application and SMS and email communication.
The designs were reviewed with product teams, who selected the version to be tested.
I then planned and conducted usability testing for both mobile and desktop versions. I prepared research scenarios, participant requirements and Figma prototypes, moderated the sessions, analysed the results and presented recommendations to the product team.
Based on the findings, I updated the final designs. However, not all research recommendations could be implemented.
One important usability issue was caused by another existing functionality in the loan process. An automatic financial analysis already used similar messages to the new Manual Decision functionality. As a result, customers could feel that their application required two separate financial analyses, which could discourage them from continuing.
Changing the existing functionality was not possible because of its complexity and limited resources. Despite the identified issue, the solution was approved for implementation due to project delays and business pressure.
Before implementation, the final designs were also reviewed by the legal and compliance teams to ensure compliance with Polish law. Any required changes to the wording were then reviewed again by the plain language specialists.
/ Deliver
I handed over the final designs to developers and explained how the functionality should behave and interact. I also provided documentation for the design systems used.
After implementation in test environments, I worked with testers to identify bugs and verify that the solution matched the designs.
Another important part of delivery was defining product and business metrics. Together with the team, we defined business goals and hypotheses, mapped the loan sales funnel and its touchpoints, and identified metrics for the new functionality.
I prepared the documentation together with the analytics team, which allowed the required events to be implemented. Analysts then created a Power BI report showing the impact of the solution on loan sales and the number of applications processed through Manual Decision.
The solution also required changes to the sales application used by customer advisors. During a business and analytics workshop, we defined possible application statuses resulting from Manual Decision and described them in user stories.
Based on these stories, I designed the required messages and statuses in the advisor application. The system was developed by several designers, and my work had to follow its predefined component library.
Because of the application’s complexity, some parts were designed directly during workshops with analysts. This helped us better understand the statuses and scenarios that needed to be represented.
At the end of the project, I also created clickable prototypes for advisor training, prepared functional descriptions and recorded training videos showing the extended cash and consolidation loan processes.
/ Summary
The Manual Decision project was a complex challenge that required me to design experiences for several different user groups, including retail customers, credit analysts and customer advisors.
Working with a Service Designer introduced me to Service Design principles and tools. Mapping the loan application process with Service Blueprint was particularly valuable and became useful knowledge for later projects.
The project also gave me extensive experience working with an interdisciplinary team, including business stakeholders, analysts, developers, testers and legal and compliance specialists.
I was involved throughout the complete design process, from discovery and research through service design, user flows, UI design, usability testing and implementation.
The project also showed the importance of making compromises in real-world product development. Despite identifying a significant usability issue during testing, the solution was implemented because changing the existing functionality was not possible within the available resources and project timeline.
