RFP — Delegator UX — Retrospective
Introduction
The original goal of this RFP was to perform an analysis of the Explorer’s Delegation UX in comparison to the staking and delegation UX of major platforms. To this end, we intended to do teardowns of six platforms (decentralized and centralized), as well as of the Explorer. We then intended to identify gaps between these platforms and Explorer and convert them into prioritized backlog items for future implementation.
All of these intentions were fulfilled, along with a few bonus additions we made along the way, and we belve we concluded with a strong narrative for improvement, and a long list of backlogs for contributors to implement. All backlogs were scored by severity and priority, ranked, and sorted into Quick Wins and Follow-on RFPs. While a single high priority list is sufficient to meet the deliverables of this RFP, we also included both a medium and low priority backlog list, which could be moved up in priority throughout implementation rounds.
Commitments Delivered
Milestone 1
Delivered
- Validate teardown methodology
- Finalize teardown exchange list
- Confirm teardown access constraints
- Decide specific scenario for each exchange in teardown set
- Finalize competitive teardown comparison matrix
- Research agentic delegation UX
Adjusted
A ticket that was added post-proposal had to do with researching agentic delegation. This addition was recommended during the proposal pitch to the community, and ended up being a very interesting task to include.
The audit also led us to add The Graph as an additional platform to analyze against the Explorer, as it has user requirements that are actually more similar to the Explorer than all other exchanges.
Deferred
Nothing was deferred.
Milestone 2
Delivered
Adjusted
Nothing was adjusted.
Deferred
Nothing was deferred.
Milestone 3
Delivered
Adjusted
Nothing was adjusted.
Deferred
Nothing was deferred.
Milestone 4
Delivered
Adjusted
To close the loop on our agentic delegation research, we validated and summarized our findings through a comment on an existing ticket related to agentic use of the Explorer: Make the Livepeer Explorer AI/agent-readable · Issue #714 · livepeer/explorer · GitHub
Also, due to feedback on our Design Brief, we decided to include one light user interview on the current Delegator UX. User interviews were consciously left out of this RFP due to the large scope and effort that is generally required for it, but we decided to include one at last minute because of how useful it might be in validating our research. As it turned out, this participant did point out the main pain points we summarized in our Design Brief.
Deferred
Nothing was deferred.
Delivered Beyond Commitments
- Agentic delegation research
- Bonus user interview
- The Graph teardown and gap specification
Commitments Not Delivered
None.
Impact
- A lucid picture of where the Explorer currently falls short with Delegator UX
- List of evidence-backed, high priority Quick Wins and Follow-on RFPs
- 12 additional backlog items (medium and low priority)
- User interview validation
- Groundwork for future agentic delegation UX
Key Learnings
- Delegation is hard to find; many visitors may never know the opportunity exists
- Orchestrator selection is intimidating, and there is no fast path to learn how to properly make a decision
- Trust and transparency are almost non-existent in the current UX. The delegation widget needs FAQs and links to risk, rather than relying on an almost hidden docs link. There is also no review step, making the movement of large amounts of LPT feel too sudden
- It’s not obvious where you are supposed to monitor your delegation; it’s very likely returning users would like to see the most important reward information right on the home page, rather than having to go to the My Account page
- There is no warning about the unbonding period until after undelegation, which very likely hurts trust between users and Livepeer overall
- Overall, the methodology turned out quite successful, but a few user interviews would likely make conclusions feel more validated
Conclusion & What’s Next
In our view, the most significant part of this analysis is the multiple layers of validation it provides for the intuition that “delegation could be improved.” In the first place, a baseline teardown of the Explorer immediately revealed the difficulty in finding the delegation widget and the lack of trust built around performing the actual delegation. Teardowns of other exchanges and protocol dashboards emphasized these shortcomings even more due to the fact that virtually all of them had the uncovered Explorer gaps completely covered. Finally, our user interview provided additional validation of where Explorer delegation falls short and confirmed that the teardown had identified the most important pain points.
One thing we would potentially change if done over is providing more user interviews, at the expense of removing a few exchange teardowns. Or request a higher budger to increase the scope. Most teardowns revealed the same things across platforms, and our one user interview turned out to be much more insightful and visceral than expected. Seeing a user’s emotional reaction to the Explorer’s shortcomings hits differently than reviewing it yourself. It would be a good idea to include a few user interviews for any larger Follow-on RFPs, such as the First-class Delegation Widget and Redesigned Homepage to Include Earn Details, as a way to illustrate and confirm how much the UX has improved.
Overall, we believe this analysis validates intuitions many may already have had, extracts less obvious shortcomings, and provides a clear, evidence-based path forward.