Continuous Discovery - Staying Aligned with User Needs
Introduction
In the fast-paced world of machine learning, staying ahead of user needs and industry trends is critical for any platform's success. At Trove, we faced the challenge of maintaining alignment with user expectations while ensuring our platform evolved to keep pace with advancements in machine learning. This case study highlights our approach to establishing a continuous discovery program that keeps developers in tune with the "why" from a user perspective, fostering innovation and maintaining competitiveness.
Situation
Trove's platform was initially successful, but as user needs and machine learning technologies evolved, we struggled to keep up. Many teams at Apple began using third-party tools for their ML data needs, expressing concerns that Trove wasn't evolving quickly enough. The Lead Engineering Manager emphasized the need for a structured approach to gather insights and prioritize features, ensuring our platform remained valuable to users.
Task
I was tasked with creating a virtuous cycle that would allow developers to consistently understand our "why" from the user perspective. This program aimed to regularly gather user feedback, track industry trends, and identify opportunities for innovation. The goal was to establish a sustainable process that would keep the platform in sync with the rapidly evolving landscape of machine learning and user needs.
Action
Stakeholder Alignment and Planning:
Initial Meetings: Conducted meetings with key stakeholders, including data scientists, engineers, ML engineers, product legal, and researchers, to gather initial input and understand their perspectives on the continuous discovery program.
Defined Objectives: Established the program's objectives, such as improving user satisfaction, increasing adoption rates, and staying ahead of industry trends.
Implementation Plan: Developed a comprehensive plan for implementing the continuous discovery program, including key milestones and deliverables.
Setting Up Discovery Channels:
User Feedback Mechanisms: Implemented various channels to gather continuous user feedback, including surveys, feedback forms, and regular user interviews. Established a partner advisory board consisting of users and key stakeholders from teams with upcoming ML work.
Usage Analytics: Collaborated with the Lead Engineering Manager and Engineering Manager partner to create a weekly metrics review session, where developers presented and explained key metrics.
Market Research: Created a monthly event called “The Future of Trove,” an open forum for presenting work, asking questions, and suggesting features. Platform developers and stakeholders were invited to attend, with PMs/EMs always present. Included live coding demos.
Prioritization and Action:
Product Backlog Updates: Regularly updated the product backlog based on insights from the continuous discovery program, ensuring high-priority items were addressed in upcoming sprints.
Feedback Loop: Established a feedback loop to inform users about how their input was being used, building trust and encouraging ongoing engagement.
Continuous Improvement:
Program Review: Periodically reviewed and refined the discovery program to ensure it remained effective and responsive to changing needs.
Internal Feedback: Solicited feedback from internal teams on the program’s effectiveness and made necessary adjustments to improve its impact.
Result
Enhanced User Satisfaction: Regularly gathering and acting on user feedback led to features that better met user needs, increasing user satisfaction and retention.
Better Alignment: The structured discovery approach ensured all teams aligned with the platform's strategic goals and user needs, improving collaboration and focus.
Efficient Prioritization: The prioritization framework ensured resources were allocated to the most impactful projects, enhancing development efficiency and effectiveness.
Sustainable Process: The continuous discovery program established a sustainable process for ongoing learning and adaptation, keeping the platform relevant and competitive long-term.
Motivated Team: The team reported feeling more connected to the “why” behind their work and looked forward to hearing direct feedback from partners and peers.
Key Takeaways
Stakeholder Engagement: Early and continuous stakeholder engagement ensures that the discovery program addresses real needs and leverages diverse perspectives.
User Feedback Channels: Diverse feedback channels provide a holistic view of user needs and experiences, driving meaningful improvements.
Regular Reviews: Regular program reviews and adjustments ensure the discovery process remains effective and relevant.
Transparent Communication: Transparent communication builds trust with users and internal teams, fostering a collaborative environment.
Implementing a continuous discovery program has been transformative for Trove, aligning our efforts with user needs and industry trends. This case study illustrates how structured discovery processes can drive innovation and maintain platform competitiveness.
