Optimizing identity verification to increase activation rates
Company
Kraken
Role
Senior Product Designer
Timeline
Q3-Q4 2023
Context
Kraken must verify users identities before they can start trading to comply with AML regulations and prevent fraud within the platform. This is often done by confirming user’s personal identifiable information (PII), verifying their identify with a government-issued photo ID and confirming their identity with a face photo.
As the lead designer for this project, I worked closely with XFN (Engineering, PMs, Content, and UXR) to optimize the verification flow by exploring ways to improve the document upload experience and increase name/address match.
Problem
Users were failing identity verification when signing up
When signing up for Kraken, users were struggling with identity verification. There was a 70% drop off from account confirmation to verification. This caused potential customers to abandon the process, impacting market growth. This also risked damaging the platform’s reputation and trust. Optimizing this experience was essential for the business.
Why was this occurring?
Through user testing interviews, I discovered that the main causes of identity verification failure was due to three main issues.
Name and address mismatch
We noticed a significant amount of users who had failed name/address mismatch, which was a critical step in identity verification.
Photo ID failure
A large number of users experienced photo ID failure. Users confused about which photo ID to upload, which files were valid and would have issues with quality of ID capture.
User errors
Users were unsure about the status of their verification and didn’t have clarity about how to fix errors during photo ID upload. There was a lack of system feedback.
Snapshots of the old ID upload experienceSolution
More guidance, real time system feedback and optimized patterns
My goal was to increase user clarity by:
Enhancing guidance through copy, notifications and patterns
Surfacing client-side system feedback by building out full stack infrastructure for error messages
Increasing PII form submission and name/address match by building address autocomplete feature
Enhanced Guidance
By optimizing patterns, optimizing components to have more visual guidelines, and adding responsive systemic feedback, we were able to guide users during the ID capture experience.
I added more guidance to help users check their ID was meeting success criteria before submitting.
Address Auto-complete feature
Our team built out the ability to auto-complete user’s addresses within the PII form by leveraging Google’s API. This reduced user errors when users were filling out the PII form, made the form quicker to complete, and increased our name/address match.
Refreshed UI, more robust Components
I built new, visually appealing components that aligned with modern design standards, ensuring that the interface was not only functional but also engaging. By adding real time guidance to the photo capture experience, we were able to give users more guidance about how to successfully capture a valid photo. Additionally, we built the infrastructure to serve more specific error messages, so users could understand why something went wrong and correct it before proceeding.
Overlaying visual guidelines and incorporating real time feedback for webcam capture helps users adjust their ID within the webcam capture.
I optimized components to add guidance during live webcam capture. This included visual guidelines for where to place your ID or face in the camera.
Capture flow
I also redesigned the document capture flow for users to upload their ID docs using their camera. I leveraged the same visual cues and step-by-step instructions into this flow to make it cohesive and offer the same guidance.
Upload flow
The redesigned upload flow simplified the process, making it intuitive for users. Clear visual cues and step-by-step instructions were incorporated to guide users through uploading their Photo IDs. We added guidance to help users successfully upload a successful ID.
Impact
Increased verification rates
We saw an increase in verification rates across desktop and mobile.
14% increase in 7-day trailing verification rates
30% increase in document submission
8% decrease in front-end errors for invalid file types
25% decrease in support tickets related to ID uploads