An app to optimise
loan recovery
An app to optimise
loan recovery
Collect Express by Capri Global Ltd · Android · 2024
Collect Express by Capri Global Ltd · Android · 2024
About Capri Global
Capri Global is a Non-Banking Financial Company (NBFC) that provides both secured and unsecured loans, including gold loans, MSME loans, and construction finance.
Brief
As a listed company offering loans, Capri deal with defaulters (who are unable to pay back on time), Capri has a dedicated “collection recovery agents team.”
To support this team, they developed an app called Collect Express, specifically designed for recovery agents who go out to collect payments. Now they asked us to redesign this app.
Why Redesign?
Why Redesign?
Poor Visual Design
Poor Visual Design
The app looked outdated and lacked visual appeal, so they requested a redesign to improve its aesthetics.
Low Engagement
Low Engagement
Agents weren’t using this new app, they still prefer to use their old systems to track incentives and report visits.
High Learning Time
High Learning Time
Due to its complex workflows, the app required an extensive learning period, which stretched to nearly three months.
Team and Timeline
Team and Timeline
2 Senior Product Designers, 1 Product Manager and 3 developers. This was done in approximately 3 months, which includes 1 month of production readiness.
2 Senior Product Designers, 1 Product Manager and 3 developers. This was done in approximately 3 months, which includes 1 month of production readiness.
Target Audience
Core Users are field agents who will be going for customer visits. 99% of the agents are male.
Core Users are field agents who will be going for customer visits. 99% of the agents are male.
Contextual Studies
We visited their Pune office and stayed for a few days. During our stay, we accompanied three agents and two managers.
We visited their Pune office and stayed for a few days. During our stay, we accompanied three agents and two managers.
Dependency on Other Apps
Dependency on Other Apps
We observed that at various stages of their workflow, they relied on WhatsApp. For instance, they used it to receive daily customer lists, send updates to managers, and confirm payments to customers.
We observed that at various stages of their workflow, they relied on WhatsApp. For instance, they used it to receive daily customer lists, send updates to managers, and confirm payments to customers.
Overuse of Technical Jargon
Overuse of Technical Jargon
When asked how they tracked their incentives, an agent explained, “We receive daily updates about earned incentives on WhatsApp.” Although the app had a dedicated incentive section, agents found it confusing due to unclear terminology and insufficient explanation of how their actions contributed to their incentives. This created a steep learning curve for understanding the system.
When asked how they tracked their incentives, an agent explained, “We receive daily updates about earned incentives on WhatsApp.” Although the app had a dedicated incentive section, agents found it confusing due to unclear terminology and insufficient explanation of how their actions contributed to their incentives. This created a steep learning curve for understanding the system.
New Usage Patterns
New Usage Patterns
When we inquired about visiting a customer using the app, the response was surprising. The actual process agents followed differed significantly from the app’s intended design. Due to usability challenges, they had adopted entirely new patterns of usage.
When we inquired about visiting a customer using the app, the response was surprising. The actual process agents followed differed significantly from the app’s intended design. Due to usability challenges, they had adopted entirely new patterns of usage.
Broken and Missing Workflows
Broken and Missing Workflows
While observing their workflows, we found that in some cases—such as leave scheduling or expense reporting—agents reverted to using older systems. These flows were either missing or inadequately implemented in the new app, leaving them no choice but to rely on familiar tools.
While observing their workflows, we found that in some cases—such as leave scheduling or expense reporting—agents reverted to using older systems. These flows were either missing or inadequately implemented in the new app, leaving them no choice but to rely on familiar tools.
Suggestions and Improvements
Suggestions and Improvements
• Add functionality to send updates directly to managers within the app.
• Enable the app to send payment receipts directly to customers.
• Add functionality to send updates directly to managers within the app.
• Enable the app to send payment receipts directly to customers.
Conclusion
Agents expressed frustration with the app’s poor experience. While it allowed them to visit customers, the process was unnecessarily long and tedious. The incentive screen lacked clarity, often failing to display accurate numbers, forcing agents to depend on their managers for updates. Ultimately, the agents used the app only because they were mandated by the company.
Agents expressed frustration with the app’s poor experience. While it allowed them to visit customers, the process was unnecessarily long and tedious. The incentive screen lacked clarity, often failing to display accurate numbers, forcing agents to depend on their managers for updates. Ultimately, the agents used the app only because they were mandated by the company.
Simplify and streamline workflows to reduce dependency on external tools and improve user engagement.
Simplify and streamline workflows to reduce dependency on external tools and improve user engagement.
How will we measure the success?
How will we measure the success?
Decrease in the time required to schedule a customer visit
Reduction of the learning period to 2 weeks
Increase in incentives earned per agent
Improved user satisfaction levels
Decrease in the time required to schedule a customer visit
Reduction of the learning period to 2 weeks
Increase in incentives earned per agent
Improved user satisfaction levels
Final Design
The app features two core workflows, and I will provide a brief overview of both.
The app features two core workflows, and I will provide a brief overview of both.
WORKFLOW 01
WORKFLOW 01
Visit a Customer
The agent’s main task is to visit customers to remind them of payments or collect them. To do this, agents go through several steps: selecting customers, planning routes, setting a daily target, and finally visiting the customers. While this may seem straightforward, the process has many underlying complexities.
The agent’s main task is to visit customers to remind them of payments or collect them. To do this, agents go through several steps: selecting customers, planning routes, setting a daily target, and finally visiting the customers. While this may seem straightforward, the process has many underlying complexities.
To simplify, I’ve broken down the workflow into 04 sub workflows to highlight the core actions.
To simplify, I’ve broken down the workflow into 04 sub workflows to highlight the core actions.
-SUB WORKFLOW 01
-SUB WORKFLOW 01
Select Customers
Old - Complex Entry Point
The agent’s main task is to visit customers to remind them of payments or collect them. To do this, agents go through several steps: selecting customers, planning routes, setting a daily target, and finally visiting the customers. While this may seem straightforward, the process has many underlying complexities.
The agent’s main task is to visit customers to remind them of payments or collect them. To do this, agents go through several steps: selecting customers, planning routes, setting a daily target, and finally visiting the customers. While this may seem straightforward, the process has many underlying complexities.
New
Clear Entry Point: Adding a button denoting a clear entry point to start their day.
Once the agent lands on the next screen, they will have a list of customers assigned by their manager. This is based on an insight we found in our research.
Everyday morning managers send a list of customers to their agents on Whatsapp which they have to visit for the day. So we suggested the team to create a feature on the managers web app that will enable them to directly assign the customers to the agent hence avoiding usage of any third party app.
Yesterday’s Overview: An agent’s tasks for the day are dependant on how they performed yesterday. As their incentive is calculated on daily basis based on the money they have collected from their visits.
New Card Design (Customer Info): We simplified the card design by just focussing on the key information that was required for an agent to take an action.
Clear Entry Point: Adding a button denoting a clear entry point to start their day.
Once the agent lands on the next screen, they will have a list of customers assigned by their manager. This is based on an insight we found in our research.
Everyday morning managers send a list of customers to their agents on Whatsapp which they have to visit for the day. So we suggested the team to create a feature on the managers web app that will enable them to directly assign the customers to the agent hence avoiding usage of any third party app.
Yesterday’s Overview: An agent’s tasks for the day are dependant on how they performed yesterday. As their incentive is calculated on daily basis based on the money they have collected from their visits.
New Card Design (Customer Info): We simplified the card design by just focussing on the key information that was required for an agent to take an action.
-SUB WORKFLOW 02
-SUB WORKFLOW 02
Route Optimisation
Old - Sub optimised route optimisation
After selecting customers the agent have to select the most optimised route to cover all the customers. Now this has some basic issues like all the routes just looks same or I can’t see the details of the route
After selecting customers the agent have to select the most optimised route to cover all the customers. Now this has some basic issues like all the routes just looks same or I can’t see the details of the route
New - Simple route optimisation
In the new design, we display the most optimised route by default. We also added a button to change sequence. As during our research, we observed that agents often need to visit a specific customer first. Based on the selected sequence, the system automatically updates to the most optimised route.
In the new design, we display the most optimised route by default. We also added a button to change sequence. As during our research, we observed that agents often need to visit a specific customer first. Based on the selected sequence, the system automatically updates to the most optimised route.
-SUB WORKFLOW 03
-SUB WORKFLOW 03
Daily Targets
Old - Unnecessary data collection
In the previous flow, agents were required to manually set daily targets to proceed with their tasks. This added cognitive load, as they had to enter this information based on assumptions, creating pressure to meet these self-set targets later on. In some scenarios customers just make it impossible to reach these targets on particular days.
Apart from this when we asked the business team why this was there in the flow they told us that this was the part of an internal study. This was done in the beginning to understand how these agents set targets for themselves. On digging deeper we realised that this was more of a data mining activity rather than a use from an agent’s standpoint. Hence we suggest them to remove it as it will ease the process or keep it as an optional step.
In the previous flow, agents were required to manually set daily targets to proceed with their tasks. This added cognitive load, as they had to enter this information based on assumptions, creating pressure to meet these self-set targets later on. In some scenarios customers just make it impossible to reach these targets on particular days.
Apart from this when we asked the business team why this was there in the flow they told us that this was the part of an internal study. This was done in the beginning to understand how these agents set targets for themselves. On digging deeper we realised that this was more of a data mining activity rather than a use from an agent’s standpoint. Hence we suggest them to remove it as it will ease the process or keep it as an optional step.
-SUB WORKFLOW 04
-SUB WORKFLOW 04
Customer Visit
Old - Complex Architecture
Once an agent select customers and a route they land on this screen. Where they can see the customers they have to visit for the day. This screen has some core issues. They can’t interact with the map on this screen, on tapping it will take you to google maps.
Once an agent select customers and a route they land on this screen. Where they can see the customers they have to visit for the day. This screen has some core issues. They can’t interact with the map on this screen, on tapping it will take you to google maps.
New - Simplified Architecture
An agent I have to visit one customer at a time. Hence we decided to focus on the customer that the agent have to visit. Once they tap on see directions the can see the detailed route to the customer’s location.
An agent I have to visit one customer at a time. Hence we decided to focus on the customer that the agent have to visit. Once they tap on see directions the can see the detailed route to the customer’s location.
WORKFLOW 02
WORKFLOW 02
Incentives
To provide context, this flow relates to agent incentives, which are a key motivator for them. This is another essential part of the app’s functionality.
To provide context, this flow relates to agent incentives, which are a key motivator for them. This is another essential part of the app’s functionality.
To simplify, I’ve broken down the workflow into 02 sub workflows to highlight the core actions.
To simplify, I’ve broken down the workflow into 02 sub workflows to highlight the core actions.
-SUB WORKFLOW 01
-SUB WORKFLOW 01
Naming Conventions
Old - Low visual appeal
Firstly, the terminology here needs adjustment. “Earnings” refers to the fixed monthly amount agents receive, while “Incentives” are additional, performance-based rewards. The current screen design is too basic and lacks the visual appeal needed to engage and motivate agents to take action. Additionally, the screen doesn’t clearly communicate the different types of incentives or the specific parameters that impact incentive calculations. This lack of clarity may leave agents uncertain about how to maximise their rewards.
Firstly, the terminology here needs adjustment. “Earnings” refers to the fixed monthly amount agents receive, while “Incentives” are additional, performance-based rewards. The current screen design is too basic and lacks the visual appeal needed to engage and motivate agents to take action. Additionally, the screen doesn’t clearly communicate the different types of incentives or the specific parameters that impact incentive calculations. This lack of clarity may leave agents uncertain about how to maximise their rewards.
New - Gamification
To boost engagement, we introduced a badge system where agents earn higher badges as their earnings increase. We then outlined the different types of incentives agents can earn, followed by a detailed breakdown of the parameters that determine incentive calculations. Each parameter has its own visual cue on progress bars to clearly indicate progress for each incentive type. As few of these parameters has penalties.
To boost engagement, we introduced a badge system where agents earn higher badges as their earnings increase. We then outlined the different types of incentives agents can earn, followed by a detailed breakdown of the parameters that determine incentive calculations. Each parameter has its own visual cue on progress bars to clearly indicate progress for each incentive type. As few of these parameters has penalties.
-SUB WORKFLOW 02
-SUB WORKFLOW 02
Parameters
Old - Outdated design
They are using some domain specific terms which was hard for the agents to understand. Also the screen lacks clarity on how it contributes to the incentive. And what happens if I miss a particular bucket.
They are using some domain specific terms which was hard for the agents to understand. Also the screen lacks clarity on how it contributes to the incentive. And what happens if I miss a particular bucket.
New - Simplified Design
In my research, I noticed that agents struggled to understand domain-specific terms, even when explained by their managers. To address this, I added an option where definitions of specific terms can be accessed for clarity. Additionally, each “Normalisation” parameter carries a penalty, which was previously unclear. I included an icon to clearly indicate the target needed to avoid penalties.
In my research, I noticed that agents struggled to understand domain-specific terms, even when explained by their managers. To address this, I added an option where definitions of specific terms can be accessed for clarity. Additionally, each “Normalisation” parameter carries a penalty, which was previously unclear. I included an icon to clearly indicate the target needed to avoid penalties.
My Experience
Overall, this project was a refreshing experience. Despite the need to move quickly due to time constraints, we ensured each step was taken thoughtfully.
In a fast-paced environment, it’s easy to get caught up in solving individual problems, which can limit a designer’s ability to explore alternative approaches. Working on a large project like this reinforced the importance of regularly stepping back to see the big picture.
Frequent reviews with design peers who weren’t directly involved provided valuable insights and helped keep our perspective broad and open to new ideas.
Overall, this project was a refreshing experience. Despite the need to move quickly due to time constraints, we ensured each step was taken thoughtfully.
In a fast-paced environment, it’s easy to get caught up in solving individual problems, which can limit a designer’s ability to explore alternative approaches. Working on a large project like this reinforced the importance of regularly stepping back to see the big picture.
Frequent reviews with design peers who weren’t directly involved provided valuable insights and helped keep our perspective broad and open to new ideas.