| Role
Lead UX Researcher @ ServiceNow
| Skills
-
Research Scoping
-
Roadmap Prioritization
-
Persona / Journey Mapping
-
AI-Assisted Research Operations
| Timeline
Mar to Aug 2026
| Tools
-
Great Question
-
Zoom + Whiteboard
-
Figma
-
Qualtrics / Adobe Analytics
Note: some internal names, participant details, and figures have been generalized below to respect ongoing confidentiality.
Developer Portal: Turning One Adoption Question into a Full Site Redesign
I led end-to-end UX Research for ServiceNow, transforming a single AI tool adoption question into the research foundation for a full site redesign. Users were barely touching a newly launched tool, and I had five months to find out why.
CASE SNAPSHOT
Context
ServiceNow Developer Portal
3 studies as the sole researcher, started project March 2026
What I did
Led end-to-end UX research
Started with diagnostic research, then shifted to foundational and generative research
How
Mixed-methods + concept validation
Combined data from secondary research, heuristic evaluation , card sort, and interviews
TAKEAWAY
Low AI adoption on Developer Portal wasn't an AI problem, it was a foundational discovery and orientation problem that had to be solved first. That reframe redirected investment away from quick UI fixes and toward the foundational portal improvements.
Overview
Developer Portal is a comprehensive online resource for developers to test and build applications and integrations on the ServiceNow platform.
-
Benefits include: PDI access, learning courses, API, reference documentation, blogs, community, etc.
-
Since its launch in Oct 2025 , the portal has surfaced Build Agent, a newly launched AI tool, on the homepage to encourage users to test build applications using AI.
My Role
Sole UX Researcher leading 3 connecting end-to-end studies, from study planning and recruitment to data analysis and final delivery.
-
Investigated low product adoption by combining qualitative and quantitative data to identify root causes
-
Collaborated with Design, PM, and Engineering, working across time zones (PST & EST)
-
Presented actionable design recommendations that shaped information architecture and content strategy
-
Facilitated strategy discussions across design, product, engineering and marketing leadership
.jpeg)
.png)
Research Timeline
STUDY 1
Build Agent Adoption
Mar 2026
May 2026
May 2026
June 2026
July 2026
Aug 2026
Gather existing research and data
Signals on low adoption delivered
Launch segmentation study
Established the four-intent model
Developer Interviews begin
Near-term design strategy finalized
STUDY 2
Developer Segments & Journey
STUDY 3
IA Validation
How I Used AI
EXPERIMENTING WITH AI
I designed a custom AI workflow capability that automated research synthesis by extracting insights from repositories and generating instant summary reports.
✔ Expanded internal research visibility
✔ Drove cross-functional adoption across multiple teams
✔ Accelerated product decision-making
PARTNERING WITH AI
I treated AI as a research assistant to accelerate repetitive operational tasks
✔ Crafted effective prompts using AI to maximize outputs
✔ Generated research plans, interview guides, and roadmap
✔ Transcribed interviews and asked follow-up questions
Research Impact
My research directly mitigated engineering risk and saved development time by putting interactive prototypes in front of users and identifying critical usability flaws.
They also served as a powerful alignment tool across fragmented product teams. By establishing an intent-driven developer mental model, I successfully grounded stakeholders under a single, data-backed vision.
By anchoring product strategy in proven user needs, this research led to executive roadmap adjustments and a complete design overhaul, giving leadership the confidence to reorder feature priorities and ship a more stable future portal.
Research Approach
Problem
Despite Build Agent's prominent placement on Developer Portal, fewer than 1% of new PDI users was actually using it to build or customize applications.
.png)
Goal
Investigate the drivers behind low Build Agent adoption and identify design opportunities to inform Developer Portal's strategy.
Why this matters
As the central developer resource hub, Developer Portal provided immense strategic value by providing:
→ Risk-free prototyping without fear of corruption
→ Learning courses and certification preparation
→ Access to developer resources
→ Community knowledge base + developer forums
.png)
-
Why aren't developers using Build Agent from Developer Portal?
-
Who is visiting the site, what are they trying to do, and what resources are typically used?
-
How do we design a portal that matches developer mental models?
Research Methods per Study
Study 1: Build Agent Signals
I began by gathering existing data to understand the current state of the portal (and general sentiment of Build Agent) using CSAT data and learnings from past case studies.
From there, I conducted a heuristic evaluation across two key workflows, surfacing design violations that pointed to a deeper issue: developers weren't struggling with Build Agent itself, they were struggling to orient themselves on the portal.
Study 2: Developer Segments & Journey
Research shifted to understanding who the users are and what their journey looked like so that we can design an experience that aligned with their journeys.
I held 60-minute interviews on Zoom with 4 Developer Advocates at ServiceNow to ground teams on developer personas, tools used, and problem-solving process.
Study 3: Developer Mental Models
Now that we've gathered enough information to build an early prototype, it was time to test whether our direction actually matched how developers think and work
I scheduled 75-minute interviews on Zoom with 9 Developers and split time in each session by four activities: establishing behavioral baseline, goal prioritization, portal content sorting, and design feedback of a new homepage and resource library.
Secondary Research
.png){kind=small}
Build Agent Telemetry
.png)
CSAT Feedback
Heuristic Evaluation
SME Interviews
Search Analytics
.png)
Behavioral Interview
Goal Prioritization
Card Sorting
.png)
Concept Walkthrough
Interviews take time and budget to set up. By first analyzing our existing assets (such as prompt logs, CSAT surveys, and case studies), I was able to pinpoint high-priority elements and highlight key friction areas on the portal:
-
Fragmented navigation and poor search
-
Outdated and buried documentation
-
No clear starting point for what to do next
Stakeholders were divided on which experience level to prioritize, but had little understanding of who was actually struggling and why. Study 1 had already hinted the answer wasn't experience level at all (veteran and new developers were hitting the same walls), so Study 2 was designed to fill that gap.
It's critical that stakeholders had as much data as possible to make informed decisions, however with less than two weeks to deliver findings, there wasn't room for a separate validation round.
Stacking the behavioral interviews with prioritization tasks, card sorting, and a design feedback session into a single sitting let me move from understanding developers' mental models to validating design concepts before leadership made any decisions. It was important that stakeholders had as much data as possible to make informed decisions.
Results from interviews laying out the Developer personas
.png)
Results from interviews laying out core user intents
Satisfaction Score of Developer Portal (2024)
Heuristic Evaluation Using Tenets and Traps
Research Questions
Activities conducted on Zoom Whiteboard
Key Insights
Insight #1
Developer Tasks Drives Tool Choice
Developers don't identify with one tool or coding style, they use whatever solves the problem in front of them. Design should support movement between tools rather than assume a fixed skill level or preference.
Outcome → Designers are currently integrating a task-based structure into final redesign.
Insight #2
Examples Build Trust Before Commitment
Developers hesitate at vague, high-stakes use cases but respond quickly to concrete, low-risk ones they can picture themselves doing. Seeing real examples does more to build confidence than reading an explanation.
Outcome → The app gallery became a top priority, and engineers are now working to remove the sign-in requirement blocking access to it.
Insight #3
Speed-to-Answer Reduces Friction
When someone hits a wall mid-task, they need fast, specific answers. Prioritizing short, accessible guidance meets users at their actual point of need.
Outcome → UX team has been collaborating with internal developers to create a resource library to replace long-form courses.
Insight #4
Unified Search Enables Discovery
Most developers don't discover what they need through the portal's search, they go to Google or AI first and only land on official resources indirectly. A context-aware and unified search experience will be the differentiator.
Outcome → This finding became a core pillar of the redesign's strategy document, prioritizing unified search over one-off UI improvements.
.png)
Impact
Setting Developer Portal's North Star
Three rounds of research led to the portal's north star, a unified, intent-driven experience that meets developers where they are.
-
Validated design concepts with developers, reducing implementation risk and rework
-
Defined an intent-driven developer model and identified top developer tasks
-
Delivered concrete design recommendations adopted by the team
-
Supported decision-making on IA, onboarding flows, content messaging, and content
-
Informed stakeholders across four separate organizations around a single, cohesive product vision
Challenges I faced
After Study 1, the scope grew from a product adoption problem into a more high-level question about the direction of Developer Portal, and adjusting to it meant bringing every new stakeholder into the same conversation instead of managing them separately.
🤝 Keeping cross-functional teams aligned and informed
→ I scheduled alignment meetings and research kick-offs and it was imperative here that I invited as many stakeholders as possible so people knew who I was, what my role is, and the research I'm working on, early.
→ After interviews, I make sure to deliver recap summaries tailored to the team, with actionable insights, and follow up with the designer to see if they need any help.
→ To manage long-term expectations, I built and shared my research roadmap. This timeline balanced immediate stakeholder requests with foundational research initiatives, all anchored to the team's OKRs.
Each study came with its own tight deadline and fast turnaround, sometimes shifting last minute. Design and product teams needed decision-ready findings fast, and that pressure often meant compromising on process and taking on more than one person's workload could realistically handle.
🏃 Matching design and development sprint velocity
.png)
→ Identifying the most critical objectives that align with project goals is essential. What worked for me was building a matrix around urgency, impact, and effort for stakeholders.
→ I maintained a research repository using SharePoint and BT1, ServiceNow's product portfolio management tool, so that all teams had visibility into my study status and priorities.
→ I had to stay flexible on my research methods, refining them to adjust to shifting timelines and stakeholder needs, without compromising the quality of the insights.
Study 2 Example:
The designer needed prototypes validated within two weeks, so I pivoted from external recruiting to internal Developer Advocates, delivering faster insights about developer journeys without sacrificing feedback quality.
UX research rarely hands you certainty, it hands you signals, and it's my job to act on them anyway. Each phase presented distinct challenges: balancing competing stakeholder priorities in Study 1, asking teams to let go of an assumption they'd already sunk time into in Study 2, and managing a tight turnaround time in Study 3.
What got me through wasn't pretending otherwise, it was trusting a good process to get me closer to the truth. That mindset comes down to three things:
→ I remain composed in the face of uncertainty
→ I believe in a collaborative research process and always bring people into the conversation early
→ I'd rather move fast with a directionally right answer than wait for a perfect one that arrives too late
🤔 Navigating ambiguity
Reflection
Looking back, one of my proudest early-career achievements was earning a seat at the strategy table. Facilitating discussions for cross-functional teams, directors, and VPs allowed my insights to shape developer needs early, before decisions hardened.
What made that seat possible was how I showed up each day: arriving with high-conviction ideas, driving proactive research initiatives rather than waiting for a brief, and facilitating discussions with the presence needed to build deep trust.
