How Learning About Our Users Informs the Redesign of a DeFi Mobile App

Many crypto start-ups rely heavily on the feedback of their power users. This often becomes an echo chamber of confirmation bias. Let's take a look at how User Research helps to inform design decisions that impact the product adoption rate.

by Risiandi Jiang – UX / UI / Product Designer / User Researcher / Individual Contributor

by Risiandi Jiang – UX / UI / Product Designer / User Researcher

The Background

Cake DeFi is a hypergrowth start-up offering retail investors the ability to invest their crypto assets through a Decentralised Finance-as-a-service on its platform. The business saw explosive growth in the 4th quarter of 2021 with about 35,000 new user acquisitions who registered and passed the KYC requirement process compare to the year before. However, our Data Analytics shows, the product adoption rate fell behind.

Cake DeFi is a hypergrowth start-up offering retail investors the ability to invest their crypto assets through a Decentralised Finance-as-a-service on its platform. The business saw explosive growth in the 4th quarter of 2021 with about 35,000 new user acquisitions who registered and passed the KYC requirement process compare to the year before. However, our data analytics shows, the product adoption rate fell behind.

The Challenge

How might we...

reduce the dormancy period and improve the product adoption rate among new users to increase revenue.

Initial Investigation

As an individual contributor, my role required me to work independently. I started my investigation by reaching out to our Customer Success team. The team shared that they had been receiving a high volume of support requests from many of our new users who had trouble understanding how to allocate their assets to Staking and Liquidity Mining. This insight provided a clue to the possible reason for the poor product adoption rate shared by the Data Analytics team.

To better understand the problem areas and build a case to initiate User Research, I performed a UX Analysis to identify and document the existing UX issues that may have caused friction for our new users.

During the process, I discovered that we needed more to address the ubiquitous problems within the crypto, blockchain, and web3 space. While Nielsen’s Heuristics Standards and Schneiderman’s Golden Rules are great guides for UX Designers, I needed to contextualize my analysis better. Inspired by those two UX Heuristics, I developed a working set of heuristics based on the problems that emerge as common themes in the crypto and DeFi space to guide our design.

As an individual contributor, my role required me to work independently. I started my investigation by reaching out to our Customer Success team. The team shared that they had been receiving a high volume of support requests from many of our new users who had trouble understanding how to allocate their assets to Staking and Liquidity Mining. This insight provided a clue to the possible reason for the poor product adoption rate shared by the Data Analytics team.

To better understand the problem areas, I performed a UX Analysis to identify and document the existing UX issues that may have caused friction for our new users.

During the process, I discovered that we needed more to address the ubiquitous problems within the crypto, blockchain, and web3 space. While Nielsen’s Heuristics Standards and Schneiderman’s Golden Rules are great guides for UX Designers, I needed to contextualize my analysis better. Inspired by those two UX Heuristics, I developed a working set of heuristics based on the problems that emerge as common themes in the crypto and DeFi space to guide our design.

Heuristic Standards for Crypto, Blockchain & Web3

Misconfigured Hierarchy
When an information flow or action is either prioritized as important or deprioritized as unimportant incorrectly based on a misalignment of the user’s mental model causes cognitive dissonance.

Incomplete Clue
When visual elements or information that could impact the user’s decision-making process are missing or misrepresented thus diminishing the value of the intended action for the user.

Redundant Clue
When the presence of visual elements or information does not serve a meaningful purpose or produce additional benefits to the user causes visual clutter.

Jargon Wall
When the intended meaning is lost in translation due to the use of exclusive jargon that might be foreign to the average user who lacks specific knowledge of the subject matter.

Misaligned Mental Model
When a language or feature designed to help the user does not do its job as intended due to an incomplete understanding of the user’s mental model.

Misconfigured Hierarchy
When an information flow or action is either prioritized as important or deprioritized as unimportant incorrectly based on a misalignment of the user’s mental model causes cognitive dissonance.

Incomplete Clue
When visual elements or information that could impact the user’s decision-making process are missing or misrepresented thus diminishing the value of the intended action for the user.

Redundant Clue
When the presence of visual elements or information does not serve a meaningful purpose or produce additional benefits to the user causes visual clutter.

Jargon Wall
When the intended meaning is lost in translation due to the use of exclusive jargon that might be foreign to the average user who lacks specific knowledge of the subject matter.

Misaligned Mental Model
When a language or feature designed to help the user does not do its job as intended due to an incomplete understanding of the user’s mental model.

Dysfunctional Experience
When a feature with the ability to improve the user’s experience is missing or inappropriately designed, thus resulting in increased effort and annoyance for the user.

Visual Dissonance
When the user experiences a disparity between what they expect to see and what they actually see.

Inconsistent Application
When a style or language is applied inconsistently to product design results in a broken pattern that induces a state of uncertainty in the minds of the user.

Wasted Real-estate
When screen real estate is inefficiently used for certain elements on the screen without producing a beneficial impact on the user experience on either an emotional or functional level.

Dysfunctional Experience
When a feature with the ability to improve the user’s experience is missing or inappropriately designed, thus resulting in increased effort and annoyance for the user.

Visual Dissonance
When the user experiences a disparity between what they expect to see and what they actually see.

Inconsistent Application
When a style or language is applied inconsistently to product design results in a broken pattern that induces a state of uncertainty in the minds of the user.

Wasted Real-estate
When screen real estate is inefficiently used for certain elements on the screen without producing a beneficial impact on the user experience on either an emotional or functional level.

NOTE: This is by no means, the definitive standard and I aim to improve it over time, but it is a good starting point to guide our design in a better direction.

More User Insights Needed

Based on the initial data from the Data Analytics team and feedback received by our Customer Success team, we hypothesized that many of our new users who remained dormant are new to the crypto space. However, we lack the qualitative data necessary to validate this hypothesis.

To close this knowledge gap the business was experiencing, I planned a series of User Research activities with the goal of capturing the insights needed to address the challenges this new segment of users was experiencing and help us design a better experience for them.

Through a stakeholder interview session with our Customer Success team, I learned that they received a high volume of support requests from many of our new users who had trouble understanding how to allocate their assets to Staking and Liquidity Mining.

We hypothesized that many of our new users who remained dormant are new to the crypto space. However, we lack the qualitative data necessary to validate this assumption.

To close this knowledge gap the business was experiencing, I planned a series of User Research activities with the goal of capturing the insights needed to address the challenges this new segment of users was experiencing and help us design a better experience for them.

Phase 1: Distilling Survey Results

Phase 2: Analyzing User Testing Results

Phase 3: Crystalizing Research Insights Into Personas

Top Takeaways From Research Insights

1. Staking and Lending are most likely the gateway DeFi products for Crypto Newbs rather
than Liquidity Mining due to the complexity of the latter.

2. One of the worst frictions in the current app is poor findability which is caused by fragmented assets-related information.

3. Poor information architecture creates navigational frictions and negatively impacts the user’s ability to perform basic tasks such as making deposits, allocating, etc. with ease.

4. The current user flows are likely to confuse users because they are trying to offer users additional abilities that do not conform to the user’s mental model and contextual tasks.

5. Product information is too overwhelming with too much technical jargon, thus discouraging most users from using a product due to the knowledge gap and fear of making a mistake resulting in losing their money.

1. Staking and Lending are most likely the gateway DeFi products for Crypto Newbs rather than Liquidity Mining due to the complexity of the latter.

2. One of the worst frictions in the current app is poor findability which is caused by fragmented assets-related information.

3. Poor information architecture creates navigational frictions and negatively impacts the user’s ability to perform basic tasks such as making deposits, allocating, etc. with ease.

4. The current user flows are likely to confuse users because they are trying to offer users additional abilities that do not conform to the user’s mental model and contextual tasks.

5. Product information is too overwhelming with too much technical jargon, thus discouraging most users from using a product due to the knowledge gap and fear of making a mistake resulting in losing their money.

Now That We Have Some Customer Insights

I facilitated a few Design Thinking workshops and invited relevant stakeholders from the Products and Engineering team, as well as, a couple of designers to ideate and align on solutions framed around “HMWs”. We also discussed tech feasibility and the timeline required for each department.

I facilitated a few Design Thinking Workshops and invited relevant stakeholders from the Products and Engineering team, as well as, a couple of designers to ideate and align on solutions framed around “HMWs”. We also discussed tech feasibility and the timeline required for each department.

Applying Mental Models to Design User Task Flows

After I've gathered a spectrum of ideas from the workshops, it is time to experiment with those ideas. Together with the user insights we uncovered from the research exercise, I started rethinking ways to optimize the user task flows to reduce known frictions by leveraging the JTBD (Jobs To Be Done) framework.

Wireframing Content Structures With the Users in Mind

Now It's Time to Test the Prototype

Once I shared the design with the team to gather internal feedback and aligned on relevant reiterations, I began applying the finishing touches to the hi-fi prototype and planned another round of user testing with a different set of test participants.

My Personal Learnings

1. When working in a team, it is always important to hear and acknowledge dissenting voices. However, it is just as important to be able to help those voices identify and recognize the source of their opinions.

2. It is common for start-ups to have a low UX maturity level, and it is part of my job to elevate and constantly advocate for the users to benefit the business through education and sharing measurable design metrics.

3. Recruiting and scheduling research participants is not my favorite task throughout this process, especially when they lie through their screening process or are no-shows on the day of the session. It is a good practice to always have backup participants ready.

4. Always have some wiggle room whenever asking for a research budget for unexpected situations.

1. When working in a team, it is always important to hear and acknowledge dissenting voices. However, it is just as important to be able to help those voices identify and recognize the source of their opinions.

2. It is common for start-ups to have a low UX maturity level, and it is part of my job to elevate and constantly advocate for the users to benefit the business through education and sharing measurable design metrics.

3. Recruiting and scheduling research participants is not my favorite task throughout this process, especially when they lie through their screening process or are no-shows on the day of the session. It is a good practice to always have backup participants ready.

4. Always have some wiggle room whenever asking for a research budget for unexpected situations.

How Learning About Our Users Informs the Redesign of a DeFi Mobile App

How Learning About Our Users Informs the Redesign of a DeFi Mobile App

﻿Many crypto start-ups rely heavily on the feedback of their power users. This often becomes an echo chamber of confirmation bias. Let's take a look at how User Research helps to inform design decisions that impact the product adoption rate.

﻿Many crypto start-ups rely heavily on the feedback of their power users. This often becomes an echo chamber of confirmation bias. Let's take a look at how User Research helps to inform design decisions that impact the product adoption rate.

by Risiandi Jiang – UX / UI / Product Designer / User Researcher / Individual Contributor

by Risiandi Jiang – UX / UI / Product Designer / User Researcher

The Background

The Background

The Challenge

The Challenge

How might we...

How might we...

reduce the dormancy period and improve the product adoption rate among new users to increase revenue.

reduce the dormancy period and improve the product adoption rate among new users to increase revenue.

Initial Investigation

Initial Investigation

Heuristic Standards for Crypto, Blockchain & Web3

Heuristic Standards for Crypto, Blockchain & Web3

More User Insights Needed