Designing for Transparency and Justice
Oath
Timeline: July 2024
Role: Product Designer
Core Responsibilities: Generative User Research, UX Design, Product Strategy
Key results: Alpha release in 2024, pending metrics
Team: Founder Jonathan, Product Manager, and Lead Engineer
Tool: Figma
As the only designer at Oath (a seed-stage startup), I took on the challenge of designing a business-critical feature: the dispute moderation flow. This feature was key to launching Oath’s minimal viable product (MVP) for its alpha release in the fall and beta release later in the year.
Entrusted with substantial responsibility—which was quite intimidating at first!—I quickly became the de-facto Subject Matter Expert on the moderator experience. From scoping the project to conducting user research and delivering final, development-ready designs, I led the entire design process end-to-end.
Armed with a little bit of naive optimism, I set out to solve this problem:
How might we restore mutual trust, collective honesty, and civil justice by turning a customer service nightmare into an opportunity to strengthen collector communities?
The problem: Our product’s origin in toy trading groups
Bad experience shopping on social media leads to loss, distress and hyper-vigilance
Oath originated within toy trading communities, such as Pokémon card and Lego collector groups, where members trade items using informal agreements in private messages on platforms like Facebook or Reddit. These transactions often rely on trust, much like making “pinky promises.”
However, when disputes arise—such as receiving items that don't meet expectations—traders face obstacles like:
Lack of formal agreements between parties
Minimal protection from social media platforms
Biased treatments from payment platforms like banks or PayPal
Social media fraud has measurable cultural and economic consequences
Over $1.7 billion in social media fraud losses reported from 120 thousand cases in the U.S. in the first three quarters of 2024, reflecting a staggering 81% compound annual growth rate over recent years.
What if…?
we reimagined a new online experience with transparency and justice at its core
Product vision of Oath
Oath as a community of traders operated on trust and open communication
Oath employs a two-pronged approach to address the lack of accountability in the current social media buy/sell/trade experience. The core product acts as a figurative security officer by matching a user’s identity with their account through a photo ID, ensuring that every Oath user is a real person. Additionally, when users utilize Oath’s easy-to-use, templated contracts, the details of their exchanges are formally documented and agreed upon.
Oath is committed to maintaining a safe and scam-free platform, as users can trust Oath to verify each individual's real identity, which is secured with end-to-end encryption.
While Oath takes precautions, some trades may still go awry, as misunderstandings can occur even among well-meaning buy/sell/traders. This is where our dispute moderation process comes into play. An Oath moderator serves as a private judge who reviews the records of the transaction, determines which party is favored with legal authority, and aids in negotiating compensation.
Although our revenue model is crucial for a startup like ours, I will refrain from delving into the specifics at this time.
More context
Business critical flow for MVP launch
At the time I joined Oath, the dispute moderation flow was the key barrier to launching our MVP, which was aggressively targeted for release by the end of summer.
While this flow isn’t the ideal “golden path” for Oath—since we aim for trades to proceed smoothly without disputes—it’s critical safety net during early adoption. Establishing trust and demonstrating reliability through an effective moderation process are key to building strong brand reputation and user loyalty.
With this understanding, I immersed myself in the problem space and turned to user research to inform my design solutions. And through extensive user research and unmoderated user interviews with 5 moderators, I discovered the challenges they face and crafted solutions to help them become empowered allies, ensuring our users feel supported in every trade.
Qualitative research insights on user needs
Buy-Sell-Traders need an impartial facilitator
For every Oath dispute, three key user roles come into play: the seller, the buyer, and the moderator.
Through secondary research and analysis of countless anecdotal stories, I identified two key expectations that traders have for the platform:
Quick responses with transparency throughout the resolution process.
Fair and unbiased treatment—avoiding any reputation for favoring buyers or sellers.
These insights shaped the core principles of Oath’s moderation flow, ensuring it was designed to instill trust and confidence in all users.
Moderators are allies that Oath empowers
Moderators are our allies! Moderators are passionate collectors who volunteer their time to maintain safe trading environments within their Facebook groups or Reddit communities.
From my interviews with five moderators, I discovered the following:
A strong sense of justice: Moderators are driven by a desire to maintain fairness and order in their communities.
Limited enforcement tools: They often feel powerless to hold scammers accountable beyond issuing bans.
This understanding guided my approach to designing tools that would empower moderators, giving them the resources needed to facilitate fair and effective dispute resolutions.
3 key design solutions
After immersing myself in the collectible toy trading experience through unmoderated interviews, extended user journey mapping, and distilling the unique needs of buyers, sellers, and moderators, I had to identity a starting point for myself.
I prioritized designing the dispute investigation process from the moderator's perspective first, validating my approach with real user feedback before shaping the traders’ dispute reporting process.
The design solutions below address three critical aspects:
Component-level interactions for the investigation form.
Feature-level communication mechanisms for dispute resolution.
Product-level information architecture to seamlessly integrate these features into Oath.
Form follows function
Some flexibility required in dispute investigation
Need #1
Through unmoderated interviews, I discovered that dispute investigations in many Facebook groups are often nuanced and require significant information gathering before any judgments can be made. Moderators frequently perform tasks such as:
Requesting proof, like screenshots of conversations.
Following up with complainants to clarify their expectations.
Reaching out to the accused (and sometimes suspected scammers) to hear their side of the story.
Conducting video calls for added clarity in complex cases.
These findings highlighted the need for Oath’s dispute investigation process to strike a careful balance. The platform had to provide enough structure to streamline the workflow while remaining flexible, allowing moderators to adapt their approach based on the specifics of each case.
Progressive disclosure and opt-in design to streamline dynamic, multi-stage judging process
Solution
Forms are not the most visually exciting things to design, but functionally, they offload the complex decision-making process of dispute moderation from our moderators’ mind to conserve their energy while retaining full control to adapt to specific cases.
If i had more time, I would conduct a thorough audit of Oath’s visual design system and standardize components across the platform. But for the alpha release, my primary focus is to build out the missing elements of the core product.
The default state “No” for the minimum requirement of proof question is intentionally designed to ensure that sufficient evidence for a valid complaint is provided before any judging takes place.
And once minimum proof has been submitted (and ideally from both parties), the moderator has the option to flag any inconsistent or potentially altered evidence (see outcome #2 vs #3 below), as there have been instances of photoshopped screenshots reported by our interviewees.
The “Flag inconsistent evidence” checkbox and “Decide in favor of” judging panel both utilize progressive disclosure based on user input to enable three distinct initial investigation outcomes.
The general decision-making logic has been validated by our Mod interviewees, but the alpha release will truly test the flexibility of this form handling real disputes and provide real user feedback from the ground.
The corresponding buyer/seller view for reporting dispute is designed with the goals of:
quickly identifying and contextualizing the conflict,
encouraging succinct reporting in the text box
Facilitating easy media uploads
After translating the details of the judging process into design, it was time to zoom out and design for the interactions between the buyer, seller and moderator during the dispute process. What would a helpful back and forth between users look like?
Opening up conversation
Multi-way communication requested for dispute resolution
Need #2
A dispute, by definition, involves at least 2 people – the complainant and the moderator.
And multiple moderators of collector community I’ve interviewed repeatedly expressed the value of having both seller and buyer dispute to help them construct a cohesive narrative.
Thankfully, our tech stack was already built with communication needs in mind, so there was no surprises for our engineering team (our founder)
Consistent post and reply interaction designs
Ideation
Aligning communication component with Oath’s source of truth mental model
Currently, disputes are often brought up in either a direct message to the moderator or a public post to the general community.
When I was comparing the different options of communication mechanisms on Oath, I prioritized the perceived user value of offering transparency between all three parties and potentially among the larger group of collectors.
My design goals in offering transparency into the dispute process are to:
help seller and buyer by allowing them to see any updates on the dispute, thus fostering trust
give moderator full context into the interaction between traders to help them remain impartial,
offer the wider collector community a lesson on what kind of practice doesn’t work.
Compare composer and form
While I knew Oath’s dispute process must offer more structure than a free-form messaging app, at first it wasn’t clear if the reporting form should be a stand-alone interface like most customer support request forms today.
I decided on the design of an embedded/integrated composer (see video) as opposed to stand-alone submission form – because of the advantages of:
signifying open-communication,
enabling context referencing and
minimizing cognitive load
An activity feed was already introduced to capture the contract signing timestamps of every Oath-enabled trade, so the design challenge for me was to integrate and present users’ communication dialog in relation to other Oath status update messages.
Reply interaction iteration
After sketching out a nested replies mechanism, I decided to use the mental model of composing individual reply messages in a separate composer (figures in the right column) which aligns better with the rest of the individual activity updates on Oath and enables clearer sorting of the timeline.
Contextual, multi-state message composer for reporting, posting and replying
Solution
As the use cases of initiating a dispute report and responding to a report both benefit from a contextual composer on the same page as the Oath contract itself, I designed a dynamic multi-state message composer adaptable for dispute reporting, update posting, replying, leaving reviews and writing thank you notes.
The reporting flow has another entry point in the Oath header under “Help” to serve users who expect to find support-related actions there.
When we release the product, I’ll likely iterate the designs based on the ratio of people using the report button toggle in composer vs the report button in header
(Missing in the UIs here, but an “X” icon to exit the investigation mode is later added next to the “pop-up” icon to enable “back” button or return to the simple composer mode!)
Situating composer
Even as I was getting the features and components of the composer and investigation form all figured out for our alpha release, the long-term roadmap for Oath’s functionalities was still transforming in our founder’s mind.
I invited our founder to workshop together and clarify the user flow he envisioned for traders navigating the Oath contract page, which enabled me to design a strong (and hopefully future-proof) information architecture and place the composer strategically.
Making sense of adaptive composer hierarchically
Need #3
Information architecture sketches balancing understandability and discoverability
Ideation
Assessment of user flow and information architecture
After defining the distinct roles of the Offers tab and the Activity tab, it became clear that the buy/sell/trade-related correspondence belongs in the Activity tab!
UI layout sketches visualizing different Information Architecture models
After comparing the 3 sketches, the option where the composer is right underneath the secondary navigation tabs offers optimal discoverability for ease of use and understandability of its relationship with other activities.
Oath page > Activity > Composer
Solution
I reorganized Oath’s main interface with a clear hierarchy, situating the composer under the Activity tab, offering clear affordance of communication capability within an active Oath case and allowing initiation of the dispute investigation process.
Putting things together~
finally!
Trader view – Oath page
Trader view – Reporting Dispute
Trader view – Reporting Dispute Expanded (Contextual Composer)
Moderator view – Investigating Report
Moderator view – Investigating Dispute / Judging
Moderator view – Investigating Dispute /Judging Expanded (Contextual Composer)
Community-approved concept currently in production
Impact and Next Steps
Oath is built intentionally collaboratively with the communities we aim to serve. Leveraging relationships our founder developed with the amazing Lego moderator community (spanning the US, UK, Canada, and Jordan), we gathered early feedback to validate the design solutions for the dispute moderation experience and address potential blind spots before the alpha release.
While our team anticipates relatively few disputes (fingers crossed!), we aim for high dispute-related task completion rates, reduced resolution times, and high user satisfaction. To gather real-world insights, we have established Discord and Facebook channels for direct user feedback, which will inform post-launch improvements.
In the future, with additional funding, I would like to conduct in-depth usability testing to further refine the moderator experience! As Oath scales, I would also explore enabling collaboration features for moderators and revisiting the private note toggle I set aside earlier.
Dare to be a newbie and embrace my designer voice
Retrospective
Working on Oath has once again shown me the power of human-centric design in shaping impactful user experiences – especially in a product’s infancy!
Initially, I felt a sense of imposter syndrome worrying that my lack of personal experience as a toy collector and moderator might hinder my work. However, I quickly overcame this by immersing myself in the problem space and learning directly from users—similar to how I approach designing for specialized B2B products.
Lastly, special thanks to Jonathan for believing in my ability and fostering a culture of curiosity and compassion. I genuinely enjoy working with such a radical thinker who wants to use technology to better the world. I’m honored to have been part of Oath’s early journey.