Unified User Privileges

eBay

One of the biggest frustrations for eBay sellers is dealing with unpaid items—when buyers commit to a purchase but never follow through. It’s a major pain point that leads to wasted time, lost revenue, and unnecessary churn for all users. To help tackle this $4.3B opportunity—yes, a $BILLION dollar problem—we developed the Unified User Privileges framework, a scalable system that proactively monitors and mitigates bad actors, stopping unpaid items before they happen.

Overview

As the lead designer for the Unpaid Items team, I spearheaded the introduction of the Unified User Privileges (UUP) system with eBay’s Risk and Policy teams. Over the course of 3 months and deep collaboration with content strategy and user research, we developed a scalable framework with a phased approach that became the foundation for new risk mitigation experiences.

Stakeholders

  • SVP Chief Product Officer

  • VP Payments

  • VP Design

  • Director of Product Management

  • Sr. Manager, Financial Services and Trust

Key results

  • 5% reduction in A/B test exceeded phase 1 auctions goal (1-3% reduction)

  • Launched to 100% roll-out in H2, 2024

Business problem
 eBay’s overall unpaid item (UPI) rate was strongly contributed to by high velocity buyers whom would delay and/or intentionally not pay for items they were committed to buying, such as auction items they won. Additionally, eBay did not have a way of limiting established high-UPI rate buyers from continuing to shopping on the platform, thus eroding trust for buyers and sellers.

User problem
Users feel that they are losing trust in eBay with the increasing rate of UPI. This equates to lost time and revenue for sellers, and buyers have a perception of faulty items when they see listings reposted.

Desired outcome
1. A vetting framework to identify a user’s presumed intent to pay based on historical behaviours within eBay.

2. A transactional blocking experience that prevents “bad buys” from participating in auctions and/or making offers on eBay.

O N E

Identifying the broader system

Expanding the team’s perspective

The project was already in motion when I joined, with cross-functional workshops just wrapped up and initial design explorations underway. After diving into the existing work and synthesizing workshop insights, I realized the team was taking too linear of an approach. I saw an opportunity to zoom out and consider the entire end-to-end user experience, ensuring we weren’t just solving for one step, but creating a more seamless and scalable solution.

Observed risks
Jumping straight into problem-solving with a narrow focus can lead to a fragmented experience—what's often called "shipping your org." When teams design in silos, users end up with a disjointed journey that reflects internal structures rather than their actual needs. To avoid this, I pushed for a holistic, end-to-end perspective, ensuring our solutions felt seamless and intuitive across the entire product.

User flows were explored only at the points of risk.

Initial explorations didn’t factor in breadth of ingress points into transactions.

Aligning on the system-wide impact

With eBay’s complex, non-linear platform, I mapped out an extensive end-to-end decision tree—from initial site entry and account creation to bidding and purchase. This system-wide view helped align the Unified User Privileges (UUP) experience within the broader ecosystem, ensuring we identified dependencies and key cross-functional partners early in the process.

Creating the decision tree was foundational in exploration and gaining cross-functional alignment.

A more detailed and accurate scope

With a shared, deeper understanding of UUP’s end-to-end impact, the team was able to pinpoint exactly where policy experiences should surface.

Key outcomes

  • Refined Content Strategy: Strengthened alignment between Risk and Policy teams and the overall product experience.

  • Streamlined UX: Reduced friction by consolidating interactions and eliminating unnecessary screens, making the experience more intuitive.

T W O

Qualitative user research

Concept testing with a variety of users

To validate our hypothesis that targeted and intentional friction could deter bad actors while enabling trusted users, I teamed up with user research to conduct qualitative testing.

We built a high-fidelity prototype simulating different price points and product categories to reflect real shopping scenarios. Our testing pool included:

  • Good buyers

  • Repeat offenders

  • Occasional UPI cases (users with unpaid item marks but not necessarily bad actors)

This approach allowed us to gauge how different user segments responded to the new friction points and fine-tune the experience accordingly.

Key research insights

It’s easy to comprehend
All buyers understood the concept of unpaid items.

Emails should be suppressed
If a buyer is blocked for any reason, they don’t want shopping emails about the item they’re block for.

Buyers want to remediate
If a buyer is blocked, they want to know how to fix it if possible or way to purchase it with full payment.

Sellers should be informed
Buyers suggest that sellers should be aware of this policy change to reduce churn they may experience.

T H R E E

Designing a scalable solution

Defining a modular experience

Given eBay’s non-linear ecosystem, I needed a way to simplify the UUP concept and clearly illustrate where it integrates across the platform. By synthesizing the multiple ingress points, I identified patterns and grouped touchpoints into a streamlined flowchart.

This approach helped distinguish flexible vs. fixed elements within the framework, making it easier for teams to align on where and how the experience should be introduced.

Navigating technical limitations

One of our main experience goals was to inform users of any restrictions before they reached a critical decision point. However, due to strict SLAs on listing pages (View Item), running real-time profile background checks with Risk and Policy could jeopardize performance.

To mitigate this, we strategically placed the check behind the “Place Bid” CTA. While this was a tradeoff, it allowed us to capture a clear signal of user intent before applying any restrictions—striking a balance between seamless browsing and effective policy enforcement.

Balancing performance and risk

Given the technical constraints, we aligned on a phased approach to measure the efficacy of our solution before introducing more tiered messaging that could impact the View Item page. While an ideal experience would help mid-risk buyers remediate their blocks, we prioritized tackling the worst offenders first. This tradeoff allowed us to validate impact early and make more informed decisions on future phases of implementation.

A/B test (US) outcome

Our test results showed that the UUP block reduced Unpaid Items (UPI) by 5% in Auctions without impacting Gross Merchandise Bought (GMB). Based on the June 2024 research readout, Payments analytics determined that the experience was successful and recommended a full rollout to 100% of the US market.