Context

• My company develops a Fintech product in Vietnam. Due to the absence of a lending license, it secures funds from Lender A and subsequently provides these funds to users, incorporating an interest rate and fee.
• On a typical workday, another Product Manager informed me of his departure in the following day. After receiving the handover from him, I became the main PIC for the Manual ID Update process. 🫣

Problem Space

Job Story

• If a user’s ID card on the system had been expired, they were unable to make any transaction. Since there was no ID Card Update Flow available in the app, users must contact Customer Service for assistance.
• Each time the Customer Service team received an ID update request and successfully validated the user’s information, they created a Jira ticket, inputted the new document data, and attached the user’s selfie along with the new national ID.
• Within my company structure, only PMs or Engineers were authorized to execute data updates in the internal system and with partners. Consequently, the PM was the primary point of contact for this process.
• As the main PIC for ID Card Update - Manual Process, I was responsible for data preparation, which involved numerous steps and requireed significant effort.

High-level problem mapping

• My early days as a main PIC of this process were a struggle because it was one of the most painful experiences I had ever encountered. It took me ~ 5 minutes per ticket.
• Additionally, the average monthly ticket volume typically exceeded 200 cases, with peak months reaching double the average.
• Average handling time for ID Update tickets was approximately 2 weeks, potentially impacting the transaction rate afterward (intention of making transaction of user).

Use Case Mapping

Solution Space

Solution Ideation

Note: Evaluate in Q2.2023

Solution 2 was the optimal approach, but due to a lack of resources for development, Solution 3 was the most appropriate choice at that time.

Final Solution

• Asked Customer Service team to input data in Google Sheet & organized Selfie + ID Image in user_id folder.
• Downloaded input file & image folder.
• Built script (thanks to ChatGPT) to re-format the data & image then updated to internal system + lender.

Impact

• Reduced SLA for updating ID Cards from 15 days to 3 days, consequently increasing the transaction rate after successful updates.
• Replaced each manual step with a streamlined review process, requiring only data review and script execution.

Reflection

Fortunately, I completely removed this boring process & built the ID Card Update Flow on the app starting from October 2024. Woohoo!!!

Sometimes, the most impactful solution is not fascinating, but rather an optimized approach that enhances existing processes.