De-Risking
the 2026
Monetization Roadmap
A mixed-methods study blending qualitative ethnography with MaxDiff-style validation that surfaced the "Infrastructure Paradox", and redefined feature tiering for a multi-million-dollar engineering bet.
UX Researcher
Qual + Quant
+ Tiering Strategy
From transactional to retained.
In a highly competitive gig-economy landscape, understanding the transition from a "transactional" user (searching for work) to a "retained" user (managing work) is critical. This research was designed to identify which platform features drive long-term loyalty versus those seen as essential table stakes.
By blending deep qualitative insights with scaled quantitative validation, I provided a roadmap for premium feature packaging that aligns with user mental models, rather than internal feature wishlists.
Which features justify a premium tier?
The platform faced a challenge in identifying which new features would justify a premium subscription for service providers. We needed to:
- Prioritize a backlog of 15+ potential features based on perceived value and willingness to pay.
- Segment needs across service verticals (Childcare vs. Senior Care) to see if a one-size-fits-all premium tier was viable.
- Identify features that drive engagement after the initial hire is made.
A dual-lens framework.
I designed a mixed-methods approach to solve for both emotional depth and statistical significance, qualitative to find the question, quantitative to defend the answer.
Qualitative ethnography
- Method60-min deep-dives
- Samplen = 10
- FocusEnd-to-end journey
- LensJobs-to-be-Done
Qualitative research reveals the anxieties and aspirations that surveys often miss, the "unseen labor" of administrative management.
Quantitative validation
- Method30-item survey
- Samplen = 300+
- TechniqueMaxDiff · Forced-rank
- LensSegmentation by vertical
To de-risk a multi-million-dollar engineering roadmap, "feelings" aren't enough. Forced-ranking quantified how much more providers valued Financial Security over Profile Customization.
A universal “Pro” tier would have failed. Different cohorts required different value levers.
1,000+ data points,
40% faster.
This was the first LLM-assisted synthesis our team had run. By feeding 1,000+ qualitative data points into a secure environment (Google Gemini), we identified thematic clusters in hours rather than days, allowing the team to move into the quantitative phase one week ahead of schedule.
What providers actually feel.
Synthesis surfaced three distinct pain dimensions, engagement, economic, and psychological, each requiring a different feature lever. A fourth strategic insight reframed the segmentation problem entirely.
The Application Black Hole
Providers invest significant time in personalized applications only to receive no feedback or status updates.
- Ghosting after personalized applications
- Vague postings lacking hours, pay, specific needs
- Market saturation, a race where expertise is overlooked
Financial Vulnerability
Providers view their work as a professional business and are highly sensitive to "stolen time" or income loss.
- Unpaid cancellations, immediate income loss with no safety net
- Payment inconsistency erodes platform trust
- Platform leakage, tempted off-platform for direct pay
Safety & Credibility Gaps
Entering a stranger's home presents inherent risks that are not always mitigated by standard marketplace features.
- The "unvetted family", asymmetric scrutiny
- Verification needs, fragmented physical documents
- Liability risk, no formal contracts or insurance
Sprint vs. Match
Discovered a fundamental split in user mindsets that reframed the segmentation problem:
- The Sprint, urgent need for any job
- The Match, selective search for high-quality, safe, long-term placements
- This split mapped cleanly to vertical (Childcare vs. Senior Care)
What providers actually value.
preference score
“The Race” vs.
“The Case.”
By synthesizing the two data streams, I developed a behavioral framework that redefined our target segments. Both groups looked the same on a feature matrix; their willingness to pay diverged sharply once context was added.
The Race
- Top JTBDGet hired faster than peers
- Will pay forProfile Boosts
- Will pay forApplication Status
- Will pay forInstant Notifications
The Case
- Top JTBDDe-risk a long-term engagement
- Will pay forClient Background Checks
- Will pay forLiability Insurance
- Will pay forSecure Payment Contracts
Data-driven product impact.
Three concrete shifts in the 2026 plan came directly out of this study:
Roadmap Pruning
Quantitative data showed "AI Creative Tools" ranked in the bottom 10% of value for all segments. I successfully advocated for removing them from the 2026 roadmap, preventing feature bloat that distracts engineering.
Monetization Tiering
Proposed a "Business-in-a-Box" subscription. Providers pay for Guaranteed Payments (income stability) and Verified Client Insights (safety), the two highest-ranked pains.
AI-Integrated Synthesis
Our team's first LLM-assisted synthesis. Identified thematic clusters in hours instead of days, advancing the quant phase a full week ahead of schedule.
Why this transfers.
The mixed-methods framework I used here, combining qualitative JTBD discovery with MaxDiff-style quantitative validation, maps directly to enterprise SaaS feature prioritization and B2B subscription strategy.
The behavioral segmentation between "The Race" and "The Case" is structurally identical to how enterprise buyers segment by role, risk tolerance, and procurement authority.