Problem
Quantitative research showed us that the account opening process of our investment app was too long and hard to complete.
Solution
An easier, faster, and secure way to open an investment account using the information we already had.
Impact
Conversion increased 6x more in the first month, and our process became the default for all the company's B2C products.
Discovery
Data Analysis
Working with the Product Manager and analyzing our conversion funnel, we discovered we had a 24% break between people who started the account opening process and completed registrations.
We decided to investigate the drop's reason in order to improve our conversion rate.
Research & Problem definition
We sent a quantitative survey to everybody who dropped off in the last three months to understand why they were interested in opening an account with us, which step they stopped, and why they gave up on continuing the process.
In the answers, 1 number raised a red flag for us: 1/4 of the interviewers had given up when they had to re-send some info for us. So, we ran qualitative research to understand this better. We discovered that the process was too long and hard to complete for most of our users, primarily because if they had to re-enter some information, they had to fill the entire form again (44 fields in total).
Delivery
Ideation
Knowing the root cause for our conversion issue, we realized it was worth it to extend our deadline to deliver a better experience for the client.
Through a Design sprint with all the team, we defined our premises and developed a solution to test with our users.
The solution was based on three premises:
-
Reduce as much as we could the required fields, or already fill in the information we had throughout bureaus.
-
Separate the process into auto-save steps.
-
Give clear feedback for the user at each stage and as faster as we can.
Validation
We validated our solution by User testing with a medium-fidelity prototype. It worked fine, but we discovered a significant issue:
-
All the users had difficulty with the Selfie step. They don't understand the photo was taken automatically (which was a demand from the supplier) and took about 2 minutes to get the picture.
-
All the users thought the document's photo was automatically taken the same way as the selfie, but it wasn't.
As we can't capture the document's picture automatically or can't put a button to capture the selfie, we change the order in our step's list. We chose to bring the document's stage first because it's more common for our users to take a photo with a button than an automatic option, and we didn't have enough time to develop this feature.
I finalize all the flow documentation on Overflow and deliver our MVP to the developers.
Result & Impact
The improvement in the process was so huge that we'd increased our conversion 6x in the first month.
It was a stunning result and beyond what we were expecting.