Meghan Urisko
Experience Research and Product Management

 

 

PROJECT MANAGEMENT

​

Stakeholders:

Director of Product
Engineering team (technical lead, engineering manager)
Computational biologists (internal advisors)
External lab partners
​

Ask: How do we leverage our data corpus and computational models to accelerate biology? Specifically: How do we remove barriers to accessing these models and enable experimental biologists to run them on their own data?

​

Approach: Project management as a discovery method—embedding directly with a fast-moving engineering team to uncover product requirements, validate strategic assumptions, and coordinate across disciplines that don't typically work together (experimental biology, computational methods, product design, engineering).

​

Timeline: 6 months (conception through launch)

​

Key PM Activities

• Cross-disciplinary translation: Worked with computational biologists to translate complex analytical methods (PCA, scVI, Transcriptformer embeddings, co-embedding with CELLxGENE Census) into user-friendly workflows for experimental biologists with no coding background

• Stakeholder alignment: Coordinated across engineers, designers, computational advisors, and external lab partners to define scope, prioritize features, and validate design decisions iteratively

• Partnership development: Identified strategic opportunity to partner with scverse consortium; drafted proposal, secured leadership approval, and defined cross-team deliverables to integrate scverse tools into the platform

• Feature prioritization & sequencing: Determined MVP scope (PCA, scVI, co-embedding, label transfer from Tabula Sapiens, differential analysis) based on user workflows and technical feasibility

• Iterative validation: Embedded feedback loops with external lab partners and internal computational biologists throughout development to ensure the tool solved real workflow problems

​

Results: Shipped a no-code single-cell analysis tool that compressed the typical 8–12 month experimental-to-computational biologist collaboration timeline to approximately one week.  The product demonstrated proof of concept that removing analytical barriers accelerates discovery—a finding that shaped the broader product strategy and validated the investment in accessibility-focused tooling.

​​

Key Insight: Project management—when practiced as an active discovery rather than pure coordination—unearths product requirements and user needs that traditional research methods might miss. By embedding with the team, translating across disciplines, and iterating alongside engineering, we built something that for both computational and experimental biology.

​

​​USABILITY TESTING

​​

Stakeholders:

Product Owner
Project Manager
Product Designer
Technical Feasibility Advisor
Additional stakeholders (ADA, Design, Legal reviewers)
​

Ask: Evaluate the usability of a multi-workflow process before production launch. The initiative was high-visibility with representation from ~20 cross-functional teams, each needing to validate that the redesign would work for their use cases.

​

Approach: Multi-round usability testing with iterative refinement between rounds, followed by end-to-end validation testing.

​

Sample Size: 40 participants across 4 rounds of testing

​

Tools: InVision Prototype, Microsoft Teams, Airtable

​​

Evaluation Metrics:

• Task Success Score

• System Usability Score (SUS)

​

Results: Testing revealed workflow concerns that were systematically addressed in each round. Visual refinements were iterated continuously as part of a broader design system rollout. The final end-to-end round validated the complete redesigned experience, creating a clear handoff to production with high confidence in usability.

​

Key Insight: Breaking a complex workflow into testable phases allowed the team to address structural problems early (when they're cheapest to fix) while design elements evolved in parallel. This phased approach reduced post-launch support burden and accelerated time-to-market.

​​

​

Card Sort

​

Stakeholders:

Product Owner
Project Manager
Subject Matter Experts (2)

​

Ask: The product wanted to add granular user control over notification delivery modes. Before building, we needed to understand whether the current information architecture allows users to find specific notifications or if substantial IA changes are needed.

​​

Approach: Closed card sort to validate how well the categorization system performed and identify where users struggle.

​

Sample Size: 50 participants

​

Tools: Optimal Workshop, Excel

​

Evaluation Metric: Success Percentage: Correct matching of notifications to their existing categories

​

Results: Several category names caused confusion; users consistently misaligned notifications with these labels. The research identified which names needed renaming and which IA reorganizations would have the highest impact. The product team could then determine scope and budget impact before development began.

​

Key Insight: Naming matters more than expected. Ambiguous terminology created friction that couldn't be solved by better UI—only by renaming or restructuring. This finding shifted the prioritization: IA improvements became a prerequisite for the new feature, not an optional refinement.

​

​​​

PARTICIPATORY DESIGN

​

Stakeholders:

Product Owner
UX Lead

​

Ask: An existing feature was underutilized. Before redesign, we needed to understand: Is it actually valuable? How do users want to use it? What functionality would unlock its potential?

​

Approach: Co-visioning sessions with two groups of potential users, using participatory design to generate ideas collaboratively rather than testing existing concepts.

​

Sample Size:

20 participants (2 groups of 10)

​

Tools:

Whiteboards, sticky notes, sample scenarios, audio recordings, Miro
 

Results:

Sessions confirmed the feature's fundamental value and uncovered two distinct use cases that had previously been treated as a single use case. For each use case, participants articulated specific functionalities that would make the feature indispensable to them. Design and engineering could then approach each use case with separate visual patterns and technical cost estimates.

​

Key Insight:

Users don't always know what they want until they're invited to co-create. By bringing them into the design process early—before a prototype existed—the team avoided building a one-size-fits-all solution and instead designed for two meaningfully different needs.

​​

​

ETHNOGRAPHIC FIELD STUDY

​

Stakeholders:

Program Manager

​​

Ask: Identify organizations and leaders doing impactful work in Detroit's digital literacy space to target for grants supporting youth education.

​

Approach: Ethnographic fieldwork combining informational interviews with embedded volunteering—spending time within organizations to understand their work, values, and gaps firsthand.

​

Sample Size:  20 participants (2 groups of 10)

​

Timeline and Scope:

30+ informational interviews
150+ hours of volunteer engagement with relevant organizations
Conducted over 6 months
 

Tools: Notebooks, PowerPoint, Kumu (network mapping), whiteboards
 

Results:  Identified and documented a detailed, up-to-date ecosystem map of organizations working in digital literacy. This intelligence enabled targeted grant distribution of $400K+ to Detroit nonprofits—the highest amount awarded across 10 cities receiving grants through the program. The work generated local and national news coverage, amplifying the visibility and impact of the identified organizations.

​

Key Insight:  Deep contextual understanding—gained through time spent in the community, not just through interviews about it—reveals ecosystem dynamics that surveys and desk research miss. Trust built through volunteering opened doors and created space for honest conversations about gaps, challenges, and opportunities that wouldn't surface in transactional research interactions.