User Research and Discovery Questions

Approach (1–2 lines)
Use a lightweight analytics-first triage: identify high-impact drop-offs and high-frequency paths with simple funnel, retention, and time-on-task queries, then translate to targeted qualitative sessions.

Key metrics & example queries

• Funnel conversion rates (completion / entry): e.g., SQL for sign-up funnel

SELECT step, COUNT(DISTINCT user_id) AS users
FROM events
WHERE event_time BETWEEN '2026-01-01' AND '2026-02-01'
AND event_name IN ('visit_landing','start_signup','submit_signup','activate')
GROUP BY step;

• Time-on-task: median time between steps per user
• Event frequency & rage-clicks: count(event_name) per session
• Error rate: percentage of events with error_flag = true

Segmentation strategies

• By user value: new vs returning, paid vs free
• By acquisition: channel, campaign
• By device/context: mobile vs desktop, geo, session length
• By behavioral cohorts: users who dropped at step X, users who completed in <2min

Translate to qualitative tasks

• Recruit: users who dropped at funnel step X within last 30 days, balanced by device/channel
• Task examples: ask participant to complete the exact flow observed; use think-aloud and moderated usability to probe intent at drop point
• Probes: ask why they hesitated, where expectations differed, what terminologies confused them
• Metrics to validate: observed pain points, ability to complete task, suggested fixes
• Deliverable: prioritized list of hypothesis-driven research insights tied to analytics (e.g., “50% drop at payment — test 5 users to validate payment UX friction”)

This ties quantitative signal to focused qualitative validation, keeping research efficient and high-impact.

Approach overview (why this matters)
As a design researcher I’d operationalize personas so they guide personalization signals without hard-coding assumptions or exposing PII. The goal: usable, privacy-safe cohorts that reflect behavior and needs, are auditable for stereotyping, and measurable for real impact.

Segmentation strategy

• Start behavioral-first: cluster on product signals (goals, task flows, feature usage frequency, friction points) + attitudinal survey anchors where available.
• Use mixed-methods: qualitative interviews to label clusters with motivations and validate patterns from analytics.
• Produce soft, multidimensional personas (vectors of traits) rather than single-label buckets to avoid stereotyping.
• Implement dynamic cohorts: time-windowed cohorts (e.g., last 30/90 days) so personas evolve with behavior.

Attribution to persona cohorts

• Assign users probabilistically to persona vectors (weights across dimensions) rather than hard assignment.
• Store only cohort IDs or hashed vector summaries; avoid storing raw PII. Prefer on-device matching or server-side ephemeral IDs.
• Use propensity scores to control for confounders when measuring behavior differences across cohorts.
• Instrument events to record cohort membership and key actions for downstream analysis.

Privacy & anti-stereotyping guardrails

• Apply differential privacy/aggregation thresholds (minimum cohort size) and k-anonymity before exporting.
• Human-in-the-loop audits: cross-check persona descriptors against interview data and diversity dimensions; remove descriptors that risk stereotyping.
• Monitor for feedback loops (e.g., personalization reinforcing a narrow behavior) and add exploration noise.

Evaluation & measuring lift

• Run randomized controlled experiments with holdout and stratified randomization by persona weights.
• Primary metrics: task success rate, time-to-complete key tasks, engagement quality (retention, DAU/MAU), and satisfaction (CSAT, SUS).
• Use uplift modeling / causal inference (difference-in-differences, instrumental variables) to estimate incremental impact per persona.
• Qualitative follow-ups: moderated usability sessions with members from each cohort to validate perceived relevance and detect misclassification or bias.
• Continuous monitoring: dashboard of per-cohort metrics plus fairness signals (unequal benefits, negative outcomes).

Implementation notes

• Pipeline: analytics -> clustering -> qualitative validation -> cohort API (hashed IDs) -> personalization layer -> A/B + holdout -> mixed-method evaluation.
• Tools: event analytics, cohort service, experimentation platform, and privacy libs (differential privacy libraries).

This keeps personas actionable, empirically grounded, privacy-preserving, and auditable while enabling measurable personalization.

Situation & high-level approach
I use journey maps to surface user goals, emotions, and pain points over time, and service blueprints to reveal backstage processes, systems, and team responsibilities. Together they translate user pain into measurable cross-functional work that can be prioritized across quarters.

Example pain point → cross-functional initiative

• Pain: Customers abandon onboarding at “verify identity” step; journey map shows high frustration and drop-off.
• Blueprint reveals causes: flaky third‑party ID provider, long manual review from Ops, unclear messaging from Product.

This becomes a cross-quarter initiative: “Reduce onboarding friction” (Q1: quick fixes, Q2: systems work, Q3: measurement + expansion).

Steps I take to get buy-in

1. Synthesize evidence: present journey map + blueprint with metrics (drop-off %, time-to-complete, support tickets) and qualitative quotes.
2. Propose phased roadmap: tactical fixes (copy, retry logic) as quick wins; medium-term (replace provider / automate reviews); long-term (policy/process changes).
3. Identify stakeholders: Product, Eng, Ops, Legal, Analytics, Customer Success — map responsibilities on the blueprint.
4. Run a cross-functional workshop: align on impact, effort, risks; use the blueprint to show dependencies.
5. Define success metrics and checkpoints: conversion rate lift, MTTR for reviews, NPS lift; assign owners per quarter.
6. Follow up with a one‑page brief and prioritized backlog items to Product Manager for inclusion in multi-quarter planning.

Why this works
Blueprints make invisible org work visible; journey maps keep the user story central. Framing initiatives as phased, measurable, and low-risk early builds trust and secures multi-quarter commitment.

Goal & constraints
Reduce median request-to-insight from 8 → 3 weeks for 20 product teams while keeping quality and rigor.

High-level approach

1. Standardize intake + triage
2. Parallelize low-effort studies (surveys, moderated rapid usability)
3. Build tooling & panel to eliminate recruitment lag
4. Automate reporting and synthesis templates
5. Add capacity via 1 FTE Research Ops + 2 contract researchers initially

Process changes

• Central intake form with RICE-style prioritization; SLA: triage ≤ 48 hrs
• Triage board (weekly) to route studies: rapid (≤3 wks), deep (>3 wks), validation
• Pre-approved study packages (recruit, screener, script, analysis plan)

Tooling & templates

• Recruitment/CRM: Respondent or UserInterviews integrated with Slack + HubSpot
• Scheduling: Calendly + automated reminders
• Recording/transcription: Otter or Dovetail auto-ingest
• Templates: one-page research brief, 2-hour usability script, 1-week diary protocol, insight card + 1-slide summary

Participant sourcing

• Internal panel: build 2,000 opt-in users segmented by persona; incentives standardized
• Backup pool: vendor contracts for specialty segments (B2Badmins, enterprise)
• Recruiter playbook and automated screener forms

Hiring / contractors

• Hire 1 Research Ops lead (tools, panel, SLAs), 1 Research Program Manager (scheduling, vendor mgmt)
• 2 contract UX researchers to handle overflow and quick-turn studies
• Train 20 PM/Design liaisons in running rapid studies

KPIs

• Median request-to-insight (primary)
• Time-to-recruit, time-to-field, time-to-synthesis
• % studies delivered ≤3 weeks
• Stakeholder satisfaction (NPS) and decision-impact score
• Panel response rate & diversity metrics

Phased rollout (12 weeks)

• Weeks 0–2: Pilot with 3 teams — intake form, triage, templates, vendor contracts
• Weeks 3–6: Build panel (seed 500), implement tooling integrations, hire Research Ops
• Weeks 7–10: Scale to 10 teams, add contractors, run training, automate synthesis pipeline
• Weeks 11–12: Org-wide launch, monitor KPIs, iterate process; aim for continuous improvement cycles monthly

Risks & mitigations

• Quality drop from speed → require checklists, sampling audits, opt-in peer review
• Panel bias → rotate vendors, continuous recruitment, quota management

This plan balances automation, capacity, and standardized practices so teams can get high-quality insights within a 3-week window.

Clarify scope & SLAs

• Define intake types (discovery, usability, evaluative), SLA targets: Response (24–48h), Prioritization decision (7 days), Delivery (pilot insights within 4–6 weeks depending on scope).
• Owner: Research Ops (sets SLA), Head of Design Research (approves).

End-to-end process

1. Request submission
   • Tool: Airtable/Confluence form + Slack / MS Teams intake channel
   • Owner: Requester (Product/PM/Designer) submits required fields (goal, audience, timeline, impact).
2. Triage & respond (SLA: 24–48h)
   • Tool: Airtable automations, Slack notifications
   • Owner: Research Ops performs feasibility check, assigns priority tag, confirms scope with requester.
3. Prioritization (SLA: 7 days)
   • Tool: Jira or Asana board + prioritization rubric (RICE/Impact vs Effort)
   • Owner: Research Lead + Product partner; Research Ops prepares materials.
4. Planning & execution
   • Tool: Dovetail for recruitment & analysis, Lookback/Zoom for sessions, Notion for protocols
   • Owner: Assigned Researcher (design researcher) runs study; Research Ops handles recruitment.
5. Synthesis & delivery
   • Tool: Dovetail + slide deck; share in stakeholder sync and uploads to knowledge base
   • Owner: Researcher delivers insights; Research Lead validates.
6. Close & measure
   • Tool: Airtable + analytics dashboard (Mode/Looker)
   • Owner: Research Ops tracks SLAs and KPIs, periodic retro.

KPIs

• SLA compliance rate (percent of intake steps completed within SLA)
• Time-to-insight (median days from request to delivery)
• Impact adoption rate (percent of recommendations implemented or referenced in PRDs)

Why: This creates transparent ownership, repeatable tooling, and measurable outcomes so research scales and drives product decisions.