Anik
Home
About
Email
Invisbly
Redesigning Invisbly; a zero-party data space where people can safely & transparently earn from their personal data.
Team of One(Design)
Anik • UX Designer • UX Researcher • Product Manager
Length
Feb 2023 - May 2023
Goals + Objectives
Mitigate number & types of errors in all flows to satisfactory percentage (80%)
Have a Task flow completion rate above or within (minimum) 40% - within 3 minutes
Autovision was apart of my Graduate thesis where I worked with early-stage startup founders
to create an MVP before a investor pitch round. Much of this project required cross collaboration
between myself & the 2 founders who have a background in product management & SWE.
Tools
Context: Setting the Table
Invisbly reached out to me to redesign their existing app experience and tasked me to create 2 feature flows, giving me minimal instruction aside from their branding guidelines & typography to do so.
How might we create 2 user flows that fulfill the product requirements while staying true to the current app's business goals?
Disclaimer
This is a condensed version of the entire process, not included are personas, priority matrixes, design system management & more. For the full version please contact me directly.
The 2 user flows were based on the Invisbly's current value proposition on answering
surveys to gain points to use on premium level articles without having to pay for said articles.
Goals
Product Requirements & Design Goals
The product requirements that were given were as a followed:
- Spend their points on premium brands
- Take surveys to earn Invisbly points
Priot to starting the project, I asked myself 3 important questions to aide me in creating a design strategy;
1. How much agency do users have in how their actions?
2. How can I create a way where users' wouldn't feel forced to complete surveys.
3. How to design for people that wanted to use their data as a means to pay for things?
Feature
Feature: User Surveys
User Surveys not only give users' agency in what they are willing to share but also gives them the power to use their data in a way that benefits them.
Feature
Being sure Spending Points
While users' can earn points to access premium content, users can also decide if they want to spend their points on specific brands. This would not only encourage engagement with Invisbly but also give more agency to users' actions.
How did I begin?
I created 4 mandated features as product requirements in order to give myself constraints when developing AutoVision.
Onboarding
Multi-Platform
Product
Use Cameras
Tier Based
Subscription Plan
Gaining Domain Knowledge
Starting from Zero
I am not a car guy & I generally do not have any clue on how cars work, so this was a big challenge for me. I did have a few assumptions but my biggest assumption was that only casual Drivers or maybe Car Enthusiasts would be my core target audience for AutoVision.
Understanding & Narrowing the Scope using Stakeholder Mapping
Mapping out potential stakeholders in order to determine scope for the project allowed me to narrow down the focus while understanding how the domain works between stakeholder relationships.
Research & Synthesizing Insights
A screenshot of my research repository which can be viewed in full, here.
In order to test my assumptions, I initially conducted 5 user interviews.

However I realized down the line that I became pigeon-holed & decided to expand my interviewees to 7 to seek out SME's or subject-matter-experts.
I utilized Dovetail to aid in analyzing & synthesizing our user interviews. (It's such a good tool)
71%
Lack of Knowledge
Participants claimed that
that learning about cars didn’t
interest them or were
afraid of damaging their car.
2/7
Worked at Dealerships
These 2 individuals ended up
giving me inside information
on the retail & enterprise
side of the automotive industry.
57%
Financial Motive
For most people, financial motives
were the biggest factor towards
car maintainence& knowledge.
"We couldn't diagnose the issue with the car before we bought it"
Participants stated that prior to bringing over a car whether it was a dealership or consumers - that they wished they had known the problems before they even bought the car & how common they were.
"Selling cars is tough, who knows if a loan is approved so our probability of gaining profit. decreased"
For car dealerships, there's always a risk to putting a bid on a car in order to bring it to the lot & so there's additional financial risk when it comes to repairs & that in itself reduces the chances of loan approval on the consumer end & eventually that will affect the dealerships' profit margin.
Pivoting the Direction - Quickly
As I had mentioned earlier, I sought out SME's to give me a better grasp of the automotive industry & through those conversations, I had to drastically pivot my direction, going from a purely consumer product to a product that affected retail businesses specifically - car dealerships.
Creating Personas for Business Needs
Creating personas initially was a challenge because of how rapidly I had to change the direction of Autovision. I had a long mental block before the pivot But once I pivoted, it allowed me to create personas that would meet business needs.

The personas targeted two different audiences while splitting one audience into different segments to create a 3-tier subscription plan.
Feature Prioritization
Once we learned what our interviewees previous and current experiences were, I used the data as a way to thin of various features in order to avoid feature creep & understand how to implement high priority features.
I utilized MosCOw feature prioritization method to assist in this regard.
Conceptualizing Interactions & Flows
Mapping Out User flows
The maps served as a mental model for me whilst I constantly gathered research & feedback in each iteration.

The map below is the core pillars of the app which give me a high level overview of features & goals needed.
I had to change a lot of my flows or rework them entirely due to the change of my direction. The flows below represent that pivot.
What Did I Focus On?
While I did build out all 4 workstreams, in this instance, I will be exploring the 'Dealership' section of AutoVision.
The overall goal of the Dealership flow is allowing Management at Car Dealerships to take a look at newly acquired vehicle's common issues.
Automatic Points Balance
When you spend your hard-earned points, you're automatically notified & are delightfully updated.
Survey Card States
When users encounter a survey, if they answer correctly the card's state changes to reflect this. Same with incorrect answers as well.
Hold to Confirm Payments
Unsure if you want to spend your points balance on something? Hold to confirm allows users the time and agency to change their mind, making user errors less likely.
Process
Conducting a Usability Audit
I conducted a short usability audit where I went through the current invisbly app in order to find specific elements that could be improved upon and could be remedied in the redesign. Below are a few of my findings that I had found.
Lack of Article Filtering
The user seems to lack the control to view specific articles at a time or filter out articles that they do not want to view. Also classification of articles seem to be non-existent aside from the limited options available or from manually search itself
Confusing Usecase of Color
While I was going through the app, I was confused why the points in the card were green if I was spending my hard-earned points? Should I be happy that I used my points? Wouldn't this be more of a negative reaction?
Unfamilar Survey UI
The survey card seems as though that it lacks a bit of consistency/standards in it's UI. Users may not be familiar with it's layout from their previous experiences.
Process
Design Iterations
Home Page Navigation
Design Iterations exploring how best to condense information spatially (left & middle) on the home screen for the purpose of navigation compared to the final design(right).
Spend Screen
Design Iterations exploring what information should be shown(left & middle) on the spend screen for the purpose guiding the user to use their points(right).
Payment Modal
Design Iterations exploring different layouts of how the payment modal should look like(left & middle)& what kind of call-to-action button should give users' more agency when they spend.
Current Experience vs Redesigned.
Designing for Reactive Users
My aim for redesigning the home experience was to incorporate more agency in the users' actions while maintaining the focus of completing surveys & spending points. I also tried to adhere to Invisbly's current design guidelines whilst calling to action to reactive events by users.
Explorations of UI
Menus: Radial vs Contextual
Designing menus especially for touch-based devices is a tricky situation. I initially wanted to work with the radial menu but based on my research it would be more ideal to stick with a traditional contextual menu.

1. Muscle Memory makes radial menus tougher to use & require prior adoption.

2. Radial Menu's have a Interaction cost (click & then dragging) that a lot of people may be frustrated with or confused by.
Making Common Issues - Common Sense
The scanned vehicle screen is supposed to have multiple drop down menus that showcase different factors of said vehicle & one of those is Common Issues.

The first version of the menu had a lot of holes in the design. Based on feedback from SME's the biggest issues were;

‍1. Lack of clear direction on what to do next

2. How can they see more details on what the common issues are?

I turned each of the tabs into actual cards & added CTA buttons to give users the guidance they sought when they expanded the dropdown. I also color coded each of the cards to show the urgency level of each issue.
Each of these feature explorations required a lot of communication between myself & stakeholders in order to create a product that is aligned with business constraints while maintaining pace to reach the deadline.
Example of Managers can use AutoVision to view common issues for vehicles.
Dealerships can view maintained & unmaintained issues for vehicles & repair costs as well.
Managers can take a photo of a car in order to gain information about the car, including car parts & common issues.
What were the Results?
Useability Testing the Final Designs
The results from testing this particular flow was mixed suffice to say. I found a need for the end user (car dealership management) but the key to any good product is execution from the job executor's perspective.
The above chart details the results with 4 testers for this flow & it's a even 50/50 split on completing the task at hand as well as the error margin.
The Chopping Block
Through out the process, I had to cut out several features in order to meet the deadline but also because implementation of the feature would be too complex in a real world setting.
The biggest cut feature for the dealership flow was depth mode, which was supposed to allow for highly detailed information of a vehicle's interior & exterior that would be useful for dealerships to have.
Doing Better to get 75% Minimum
In order to increase the task flow completion rate, I would try to simplify the UI elements on the vehicle page, due to both users claiming that dropdowns were NOT obvious options to finding the common issues buttons in order to proceed to the end state.
I would also restart the project from scratch in order to implement Object Orientated UX  to determine what the key objects are, relationships between those objects. call-to-actions, & attributes in order to grasp the full scope of the project from the beginning rather than reworking & backtracking because of specific requirements/business needs.
I also revised the final screens in order to push the envelope for my own design skills & correcting my own mistakes to create a better experience for my target audience.
My own Learnings
I learned to work in unfamiliar waters by learning how to design dashboards for tablets & learning about a whole new domain while using that data to inform my own design decisions that will help with business needs.
Proposed Future
While this was a conceptual MVP for the client, I purposed that in order to measure the success of this redesign, that we should take a look into different metrics using methods such as A/B testing, surveys, heatmaps/eye-mapping and usability testing in order to determine important KPIs. My suggested KPIs to look into were churn rates, net-promoter scores, customer health scores, & customer retention level.
Metrics to success post-launch

"When you hit your lowest point, you are open to the greatest change" - Avatar Aang

Back to Home