B2B / Web
Microsoft
I led UX and shipped multiple features and vision on Azure cloud platform, enabling users to have management over cloud limits and issues for smoother cloud success. This cut cloud failure rates by 70%, reduced support volume, and tripled CSAT.
Role
Product designer
Project duration
2023 - 2025
Skills
UX/UI Design
Product Strategy
Prototyping
User research
Visual design
Team
UXR, PM, SWE
Microsoft Azure
Reshaping Cloud Quota Management UX
Year
2023 - 2025
My role
Designer
Team
Research, Engineer, PM
What's been done
Azure is Microsoft’s cloud platform that enables users to run and manage cloud resources. Quota management has long been a known pain point within the Azure experience.
In this project, I led UX strategy efforts to establish user foundation, align teams on a shared vision, and uncover design opportunities to improve the complex quota workflow.
Impacts
- Shaped a future vision (including AI opportunities) and aligned cross-functional teams
- Enabled execution of multiple UX initiatives grounded in user insights and strategy
- Increased CSAT from 25% → 75%
- Reduced quota and capacity issues by 70%
- Cut cloud workflow failures by 50%
Overview | TLDR
3,000 support tickets per quarter. The cost of invisible quota limits.
The Problem
Azure is Microsoft's cloud platform serving billions of customers. Azure was seeing 3,000+ support tickets per quarter related to quotas limits - these are cloud resource limits that customers were hitting unexpectedly, causing deployment failures and operational disruptions.
My Role
I was a product designer on a UX-initiated project, collaborating with product and engineering team to drive the design strategy from research through implementation.
Research Approach
User interviews, surveys, heuristic evaluation, system analysis
Key Gaps Discoverd
No proactive warnings
Users discover quota issues only after deployments fail and always too late to prevent errors
Portal doesn't scale
Enterprise users managing 100+ quotas lacked efficient workflows
Planning & troubleshooting challenges
Users struggled to forecast capacity needs and troubleshoot failures effectively
The Strategy
Two Tracks to Transformation
Based on those insights, I proposed a two-track approach to address immediate needs while building toward a strategic vision.
Short term solution
Feature Discovery & Onboarding
Improve tool discoverability and adoption addressing
Dashboard with Proactive Insights
Surface alerts and usage insights before failures occur
Bulk management
Enable efficient management of quotas at enterprise scale
Error Message Redesign
Provide clear guidance when issues occur
Through my research, I identified planning and troubleshooting as the top pain points. Working closely with the team, I developed the Copilot concept to provide AI-powered assistance. I validated this direction through a survey of 100+ users and collaborated with the Azure partner team on design and technical feasibility.
North start vision
AI Copilot
AI-powered capacity planning, proactive recommendations, and intelligent troubleshooting
The Solution
Never hit a limit unexpectedly again, Troubleshooting without waiting
01 Feature Discovery & Onboarding
Help users discover tools for better management
Designed a landing page that highlights key features, helping users discover management tools instantly when entering the Quota portal.
02 Dashboard
Quick-glance insights
Designed a dashboard with modular cards that surface alerts, requests, and forecasts, enabling users to monitor and respond faster.
03 Bulk management
Efficient management at enterprise scale
Enterprise users manage hundreds of quotas at scale. Designed advanced filters and saved views to enable efficient bulk actions.
04 Error message redesign
AI-powered troubleshooting and planning.
Copilot AI help users resolve any failures instantly, providing actionable recommendations and reducing support dependency, and plan for future with less chance of failures.
Why Copilot?
AI-Powered troubleshooting & planning
Through an opportunity survey with 100+ users, I learned that what users most wanted is getting instant help understanding failures and planning capacity.
User Needs
Instant answers
No waiting days for suppor tickets
Clear explanation
Not just "there is a issue" but WHY and HOW to fix
Prevent future issues
Start planning ahead
What Copilot enables
Real-time troubleshooting
Instance diagonosis when things fail
Contextual guidance
Explains root cause + offers specific fixes
Predictive planning
Forecasts needs
Impacts
70% fewer tickets, 40% fewer failures, New product direction
Established a strong UX foundation that simplified complex workflows
Drove a shift from short-term fixes to long-term solutions
Facilitated cross-functional conversations to align design, product, and engineering around shared vision
* These improvements reflect multiple factors including but not limited to quota management enhancements
The Process
Research & Insights
Rethinking the way users manage quota
Before jumping into solutions, I needed to understand the users first. I led research that uncovered target users roles and workflow, then mapped findings into a JTBD framework. This gave the team a solid user foundation, fill intheuser gaps, and revealed clear user needs and opportunities to rethink about the quota mangement exprience. Other than this also conducted heuristic evaluation of curent quota experience to find UX gaps.
PAIN POINTS
Lack of guidance
Users don't know how to resolve issues when cloud tasks fail, so they rely heavily on support.
Inefficient management
Current tools are hard to discover and don't scale for enterprise users.
Lack of visibility
Problems aren't visible until failure, leaving no opportunity to prevent issues.
DESIRED OUTCOMES
Clear guidance
Users can self-resolve issues with clear, actionable recommendations, without relying on support.
Efficient management at scale
Users can easily discover tools and manage quotas efficiently at enterprise scale.
Proactive visibility
Users can see potential issues before they become failures, enabling proactive prevention.
Goal
Help users understand, manage, and resolve quota issues independently, reducing failures and reliance on support.
Prioritization
Collarborative solution solving
I did opportunity survey to prioritize needs and ran two rounds of workshops to prioritze ideas and for solutions that are best for the users.
Ideation
Finding ways for solution
Based on the finding, I focused on three themes: guidance, efficiency, and transparency to improve quota mangements.
Short-term
Guidance
Feature discovery: Help users find management tools and drive adoption
Tutorial: Guide users to adopt features through onboarding
Efficiency
Saved filter sets: Let users save filters that match their workflows
Duplicate alerts: Streamline alert creation by duplicating existing ones
Bulk subscription: Manage multiple subscriptions at scale
Transparency
Access to past requests: Track request history and estimate resolution time
Cost transparency: Enable better forecasting with clear cost insights
North start vision
Copilot
Proactive management: Prevent issues by proactive
Error explanation: Explain issues or erros that is clear to users
Recommendation: clear guidance for troubleshooting
Ideation
Recommendation & Insights
Onboading & Learning
Management at enterprise scale
Supportability
Copilot: Proactive management
Copilot: Self-troubleshooting
