My role & responsibilities

Lead UX Researcher, I worked alongside product managers and engineers to gather user feedback, set up survey study program, recruitment, daily communications, data collection and data synthesis, weekly reporting, final report analysis.

Time Spent – Approximately 6 months.

Product Impact

20% Increase in overall satisfaction

25% Reduction in error impact

6 Hardware and software product upgrades, before GA

Product challenge

Ensuring AB4's features would be adopted seamlessly by officers accustomed to the previous version (AB3), while maintaining the critical reliability law enforcement demands.

The AB4 represented a significant evolution in our camera lineup, introducing enhanced POV accessories and bi-directional communication features. However, these advancements needed to integrate seamlessly into officers' existing workflows without compromising the reliability they depend on for evidence capture.

UX challenge

Transforming Axon's approach to beta testing. Historically, product managers gathered feedback through ad-hoc conversations, resulting in isolated insights without systematic analysis across users. This meant decisions were often based on feedback from one or two customers, without validating if issues were widespread.

Advocating for the role of UX research: Demonstrate the value of a structured, methodical approach to collecting and analyzing user feedback.

Evangelizing research: Developed a comprehensive research plan with clear goals that would enable us to track trends and generalize insights across multiple agencies.

Choose how to measure - Methodical approach

Our mixed-methods strategy deliberately distinguishes between change aversion and genuine usability issues. Through systematic data collection and analysis, we could track how officers' reactions evolved from initial resistance to substantive feedback about specific functionality.

Data Collection Framework:

• Structured weekly surveys

  • Quantitative metrics on performance and satisfaction

  • Open-text fields for detailed feedback

  • Consistent format to enable trend analysis

• Qualitative investigation

  • Semi-structured interviews with officers

  • Real-time bug reporting through mobile app

  • Weekly stakeholder discussions

Choose who to measure with: Participating agencies

Our sampling strategy targeted diversity in agency size and operational contexts. Of the 24 participating agencies, we included:

1. Diverse agency sizes (large, medium, and small departments)

2. Geographic distribution across North America

3. Various shift patterns and duty types

4. Willingness to test new features

Engagement Strategy

The program's execution required careful orchestration of multiple feedback streams while maintaining high engagement across all agencies. Through personalized communication and consistent follow-up, we achieved average response rates of 75% throughout the six-month period.

• Initial Phase (First 6 Weeks): Weekly surveys capturing initial reactions. Intensive monitoring of critical issues | Regular stakeholder synthesis meetings.

  Sustained Phase (Post 6 Weeks): Bi-weekly surveys to prevent feedback fatigue | Targeted follow-up on specific issues. Ongoing analysis of adaptation patterns.

• Personalized Follow-up: Individual email communications to maintain high response rates.

• Synthesis Meetings: Weekly stakeholder meetings to review findings and prioritize actions.

Weekly Synthesis

The weekly synthesis process became increasingly refined as the program progressed.

Each week, I would analyze the incoming quantitative data for trends, while simultaneously conducting thematic analysis of the open-text responses.

Our content analysis of qualitative feedback started revealing clear themes:

• Critical safety concerns (e.g., accidental shutoffs during encounters)

• Workflow disruptions (e.g., two-step confirmation process)

• Ergonomic challenges (e.g., button accessibility with tactical gear)

This thematic analysis helped prioritize changes based on impact severity rather than just frequency of mentions. For instance, while many officers initially reported discomfort with the new button placement, our longitudinal data showed this largely resolved through familiarization. However, issues like accidental shutoffs during physical encounters persisted, indicating a fundamental design flaw rather than change aversion.

By replicating this process across 24 agencies and comparing satisfaction, performance, and error rates, we could identify common pain points and feature requests that spanned multiple departments.

Redesign the Product - Impact and Outcomes

Over six months, the program generated several key insights, directly influencing the product's development:

• Button Design Evolution
  Before: Initial feedback suggested general dissatisfaction with button tactility, but longitudinal data revealed specific scenarios where the design created operational risks. Early quantitative data showed satisfaction rates hovering around 60%, but our qualitative analysis revealed that officers weren't simply struggling with change – they were encountering serious operational issues during critical situations.
  "The power button is very hard to manipulate and requires you to push much further in than the body 3 making it more difficult to power on."
  After: By tracking these incidents through weekly feedback and detailed officer narratives, we built a compelling case for modification. The resulting design changes, including enhanced tactile feedback, leading to satisfaction rates climbing to 85% by program end.

1. Training and documentation strategies: Longitudinal data revealed enhanced training could bridge the gap between design intent and real-world usage.

2. Across 6 months of consistent feedback helped us track user satisfaction consistently improved after key design changes, with the number of users reporting "extremely satisfied" rising from 17% in February to over 44% by June.

3. The graphs showed that as the product changes were made, the number of users reporting errors like accidental power-offs steadily declined. By the end of the beta, over 70% of users reported no significant errors, demonstrating the effectiveness of the iterative improvements.

Successes:

• Moving from ad-hoc customer conversations to systematic data collection fundamentally changed how Axon approaches product development. By implementing a structured framework, and triangulation of research methods established a new standard for how we evaluate and refine products before release.

• Building trust with law enforcement officers proved crucial to our success. By maintaining consistent communication channels and demonstrating how their feedback directly influenced design decisions, we developed strong partnerships with our users that led to increasingly detailed and actionable insights.

• The long-term impact of this program extends beyond the AB4 launch. We've established a new baseline for how Axon approaches user research in product development. The methodical approach to data collection and analysis demonstrated that user research isn't just about validation – it's a crucial tool for risk mitigation and feature optimization in high-stakes environments.

Process Improvements Identified:

• The manual nature of survey distribution and data compilation became increasingly challenging as we scaled from 5 to 24 agencies. A more automated system for data collection and preliminary analysis would have allowed more time for deeper investigation of emerging issues.

• Earlier involvement in hardware design decisions emerged as another crucial opportunity. By integrating research earlier in the development cycle, we could reduce iteration cycles and validate design choices before final implementation, potentially saving both time and resources.

• Research is iterative and should be ongoing, users don’t like change especially when they are experts within a product that they use day in and out. Any significant changes that make them relearn the basic workflows is not a satisfying experience. This realization has profound implications for how we approach feature development and implementation.

• Additionally, expanding our contextual inquiry methods could provide even deeper insights into how officers interact with our products in real-world situations. This understanding would help bridge the gap between controlled testing environments and actual field use, leading to more robust and practical solutions.