Karen Prieto.
Karen Prieto.

All projects

Onboarding redesign that increased conversion by 15% and drove a 4.8⭐ app rating.

Industry

Fintech / Identity

Company

SoyYo

Business model

B2B2C

Challenge

The app’s onboarding process — where identity verification took place — was the primary pain point in the customer journey, driving low conversion rates and the leading cause of low app store ratings.


How this was impacting the business:

  • Low conversion rate: The 26-action flow created excessive friction and led to user drop-off.

  • The low app store ratings were severely impacting the brand, affecting both user trust and B2B relationships.

  • About 90% of the customer support cases were due to usability issues during the onboarding process.

  • Attractiveness bias: Users distrusted the app, as they perceived its “look and feel” as outdated and insecure. This makes them fear data theft, hurting the product reputation, especially since it involves sensitive biometric data.

Outcomes and impact

  • Increased conversion by 15%.

  • Boosted app store ratings from 3.0 to 4.8, in both andoid and iOS.

  • Reduced customer support requests.

  • Users perceived a more clear, friendly and less intrusive experience.

  • B2B clients valued the new flows for the improved user experience and stronger alignment with their existing processes, especially when integrating them into their platforms.

My role

I led the full design process, starting with analyzing product metrics and conducting research with users, customer support, and analytics teams to uncover pain points and deeply understand the reasons behind the problem. I then redesigned key flows, information architecture, and the user interface based on insights and UX principles. I validated and iterated on the solutions through usability testing, while collaborating closely with the tech team to ensure feasible, high-impact outcomes.

Solution

1. Unifying log in / sign up flows to simplify access

🟠 The problem — before:

The flow began with users having to choose between signing in, logging in or using a QR option. This led to frequent mistakes, wasted time, and even technical integration issues within client apps.


🟢 The solution — after:

I redesigned a unified flow where, with one input, the system automatically detects if a user is registered and directs them accordingly, creating a seamless experience. This also allows companies to better integrate the flow into their processes and systems.



  1. Enhancing focus in input fields:

🟠 The problem — before:

With many fields on the screen, users get easily distracted and, due to an unusual order, they input data into the wrong field, leading to errors. 


🟢 The solution — after:

Redesigned the input form into clearer, step-by-step screens, reducing cognitive load and improving focus. Since the ID number was crucial to start the process, I added a confirmation modal to prompt users to verify it, significantly reducing input errors.


  1. Providing alternatives when errors occur

🟠 The problem — before:

Users may not receive the security code (OTP) due to messaging line failures or entering the wrong number — with no option to fix the issue. This led to significant drop-off.


🟢 The solution — after:

Enabled users to review the number they entered, correct it if need it and receive the code via SMS and WhatsApp to increase conversion.



  1. Clear and visual guidance to prevent errors

🟠 The problem — before:

The Identity verification process required a selfie but users failed due to unclear instructions for the biometric photo, often ignored, or quickly forgotten, leading to repeated mistakes.


The feedback provided during the selfie capture was unclear to users and presented accessibility issues due to low contrast.


🟢 The solution — after:

I redesigned the instructions using the "Show, don’t tell" principle — combining clear visuals with concise text to guide users and reduce errors during photo capture.


I designed all the feedback messages, used visual color cues to reinforce feedback and increased contrast levels to improve improve the success rate.


  1. Improving usability in the most challenging step

🟠 The problem — before:

ID photo capture was the weakest step for 3 main reasons:

  • Users didn’t understand its purpose, feared data theft, and often abandoned the process.

  • The instructions were unclear. Users often took photos of the wrong side, with glare or shadows, or with cut-off corners.

  • When an error occurred, they didn’t understand what they were doing wrong, leading to repeated mistakes.


🟢 The solution — after:

I added an explanation of why the ID document photo was required, redesigned the instructions using the “Show, don’t tell” principle, incorporated visuals to avoid confusion, and designed specific error messages for the most common mistakes — all to improve the success rate of identity verification.


All projects