Banco is a large private sector bank with over 2M customers, 100 branches, and 5B in deposits. The bank is currently focused on their personal loan services and offers a mobile friendly loan request process on their app. Customers need to sign up, provide financial information (steady income) and can then get access to loans based on their credit profile. Banco’s primary goals are to uncover areas of friction, explore feasible ideas and aim at increasing loan requests.
After launching their app, the Banco team decided to use UXcam to learn more about their customer in-app journey and implement solutions to render the app frictionless. They first watched several session recordings and checked the screenflow feature to better grasp how customers navigated their app. From that information, they were able to construct different funnels to analyze conversions and drop offs which included all the necessary steps from onboarding to completing a loan request.
The team constructed a user funnel for the loan request process using the steps outlined below. Choosing a user funnel allowed to track conversion by the user and not within one session since some steps such as OTP retrieval did require ending the session. For customers who are qualified for a loan, the in app steps are the following :
- Request a loan
- Fill in loan details
- Confirm loan details
- Enter the OTP code to finalise the process.
The team discovered something very alarming when analysing in more detail the loan request process. They identified 22% of drop offs came from the OTP identification in the session funnel they created. It was clear that the OTP verification was a big area of friction for the app and a problematic screen. The team decided to check to take a closer look on what was happening on the page and used the heatmap feature. They checked for rage taps (when users click repeatedly within a short amount of time) and discovered that most rage taps happened on the resend code button. Thanks to filters, the team viewed sessions with OTP verification screen and high amount of rage taps. They were quickly able to validate that users were frustrated because they had to quit the page, did not receive the code in time, and just ended up abandoning the page.
It was clearly not feasible to simply remove the OTP check as this would highly be a security risk. To address this issue, the team decided to implement an automatic OTP/SMS detection integration where customers do not need to leave the app to retrieve the code. Instead, the code is automatically filled in the required field just like the image below. They also worked on removing any bugs from their side relating to the speed of the text message. They believed this would reduce friction and allow for a smoother experience.
Once they coded in this option, Banco’s team decided to release it progressively to the users. They waited a few weeks and created a user funnel with the same steps. Within UXcam, they were able to compare funnels for the 2 different app versions using the group by app version option. As below, the team could analyse the difference in drop off rates within the 2 versions. The result was clear, the OTP changes clearly helped reduce drop outs at the OTP step by 15%.
Do you want to analyse a particular process within your app?
- Review full session recordings to grasp a better understanding on the screens and events your users make when navigating their app.
- Use screenflow to identify the top path and validate the pattern you have identified while viewing session recording.
- Track specific events by sending custom events and user properties to UXCam such as request credit card or filled in form.
- Create a funnel with the appropriate steps and identify where the most drop offs are occurring.
- Make appropriate changes and release a new version. Compare the changes within UXCam using group by app version when constructing the funnel.
- Iterate to continue making improvements.
*Name of real company has been replaced for privacy reasons
Images are examples, not real data