Use this when building user personas for a client's product, usually as part of discovery or journey mapping work.
How it works
- You provide the client context, product description, and any available research data
- The skill synthesizes the data into 3 JTBD-grounded personas with pain points, gains, and behavioral insights
- It returns client-ready persona cards with product fit assessments and open research questions
Prompt
You are building user personas for a Kate Makrigiannis consulting engagement. Kate is a fractional product leader whose discovery and journey mapping work centers on jobs-to-be-done, not demographics. Before writing, read knowledge/voice-tone-guide.md -- use the client-facing voice. These personas will be shared with the client's team.
Inputs I will provide:
- Client: {{CLIENT}} (company name and context)
- Product: {{PRODUCT}} (the product or feature these personas will inform)
- Research data (optional): {{RESEARCH_DATA}} (survey results, interview transcripts, analytics, or other user data)
- Additional context (optional): {{CONTEXT}} (engagement stage, what Kate already knows, specific focus areas)
Step 1: Gather and analyze
Read any provided research data files. If engagement history exists, check knowledge/engagement-history.md for prior work with this client. Review knowledge/pm-discovery-frameworks.md for Kate's persona and JTBD frameworks.
If research data is provided, analyze it for:
- Recurring behavioral patterns (not just demographic clusters)
- Jobs users are trying to get done and the contexts that trigger them
- Frustrations, workarounds, and unmet needs
- Surprising or counterintuitive patterns in the data
If no research data is provided, build personas from the product context and domain knowledge, but flag clearly that these are hypothesis personas requiring validation.
Step 2: Build 3 personas Frame each persona around a job-to-be-done, not a demographic profile. Demographics are supporting context, not the organizing principle.
For each persona, produce:
Persona [Number]: [Name]
Who they are
- Role, context, and key characteristics (age and demographics only if relevant to the job)
- What makes this persona distinct from the others
Primary job-to-be-done
- The core outcome they are trying to achieve, stated in their language
- When and how often this job arises
- What triggers them to "hire" a product for this job
Top 3 pain points For each:
- The specific challenge or obstacle
- How it impacts their ability to get the job done
- Severity (High / Medium / Low)
Top 3 desired gains For each:
- The benefit or outcome they want
- How they would measure success
- How well the current product (or competitors) deliver this
One unexpected insight
- A counterintuitive pattern or motivation that the team might not expect
- Why this matters for product decisions
Product fit assessment
- How {{PRODUCT}} addresses this persona's job
- Gaps between what the persona needs and what the product delivers
- Biggest risk if this persona is ignored
Step 3: Cross-persona analysis
Persona Comparison
| Dimension | Persona 1 | Persona 2 | Persona 3 |
|---|---|---|---|
| Primary JTBD | ... | ... | ... |
| Biggest pain | ... | ... | ... |
| Product fit | Strong/Moderate/Weak | ... | ... |
| Priority | ... | ... | ... |
Shared Patterns
Jobs, pains, or gains that appear across multiple personas.
Research Gaps
Questions that remain unanswered and would strengthen these personas. For each, note what research method would help (interviews, surveys, analytics review, etc.).
For mapping the customer journey for each persona, use the journey-map skill. For preparing interview scripts to validate personas, use interview-script.
Example Output
Input
- Client: Novu Health — a Series A digital health startup offering a chronic condition management platform for patients with Type 2 diabetes
- Product: A new medication adherence feature within the Novu Health mobile app that sends reminders, tracks dosing history, and surfaces personalized coaching nudges
- Research data: Exit survey from 340 app users (42% report forgetting doses "at least once a week"), 8 patient interview transcripts, analytics showing 68% of reminder dismissals happen between 11am–2pm on weekdays
- Additional context: Mid-discovery phase; Kate has completed stakeholder interviews with the clinical and product teams; client wants personas ready for a journey mapping workshop next week
Output (abbreviated)
User Personas: Novu Health Medication Adherence Feature
Hypothesis-informed personas built from patient interviews, exit survey data, and behavioral analytics. Ready for journey mapping workshop. Flagged research gaps require validation before roadmap commitments.
Persona 1: Marcus
Who they are
- 54-year-old regional sales manager; diagnosed with T2D three years ago; manages a demanding travel schedule with irregular meal times
- Distinct from other personas because his adherence failures are situational, not motivational — he wants to take his medication, but his environment works against him
Primary job-to-be-done
- "Help me stay on top of my medication without it derailing my workday or embarrassing me in front of clients."
- The job arises daily, triggered most acutely when he's in back-to-back meetings or on the road — exactly the 11am–2pm window where analytics show the highest dismissal rates
Top 3 pain points
- Reminder interruptions during client-facing moments — dismisses alerts mid-meeting and forgets to re-engage; severity: High
- No easy way to log a delayed dose — app treats a missed window as a missed dose with no nuance, creating guilt and disengagement; severity: High
- Coaching nudges feel generic — content doesn't account for travel days vs. desk days; severity: Medium
Top 3 desired gains
- Quiet, non-disruptive reminders (vibration, lock screen only) he can act on discreetly; currently no competitor does this well
- A "I'm running late" snooze that reschedules intelligently rather than marking a failure
- Context-aware coaching that acknowledges his schedule patterns; success = fewer than 2 missed doses per month
One unexpected insight
Marcus is more motivated by not looking irresponsible in front of his doctor than by health outcomes per se. His language in interviews centered on "being the kind of person who manages this well" — identity, not longevity. This suggests that social accountability features (shareable adherence summaries with his care team) could outperform clinical outcome messaging for this segment.
Product fit assessment
- Current reminders address the job but ignore the professional context where failures occur
- Biggest gap: no flexible rescheduling flow; the app's binary pass/fail logging actively discourages re-engagement after a disruption
- Biggest risk if ignored: high-functioning patients like Marcus churn silently — they don't complain, they just stop using the app
Persona 2: Diane
Who they are
- 67-year-old retired teacher; lives alone in a suburban area; manages T2D alongside hypertension and mild arthritis
- Distinct because she's managing multiple medications across multiple conditions and is the primary coordinator of her own care with limited family support nearby
Primary job-to-be-done
- "Keep all my medications straight so I don't accidentally double-dose or miss something my doctor will ask about at my next appointment."
- Triggered at the start of each day and before each quarterly clinical visit
Top 3 pain points
- App only tracks one medication — she keeps a paper list for the others, creating a fragmented system; severity: High
- Small text and tap targets — arthritis makes precise interaction difficult; she has dismissed reminders accidentally; severity: High
- No appointment-prep summary — has to manually reconstruct adherence history before visits; severity: Medium
Top 3 desired gains
- A single view across all her medications, even if Novu only manages one
- Accessible UI with larger controls and a simple confirm/undo flow
- An auto-generated "appointment summary" she can print or share with her provider
One unexpected insight
Diane trusts the app more when it acknowledges uncertainty. In interviews, she responded positively when language was framed as "based on what we've tracked" rather than definitive statements. Overclaiming accuracy triggered skepticism and reduced her confidence in the data. This has direct implications for how coaching nudges and adherence scores are written.
Product fit assessment
- Feature delivers core reminder functionality but creates friction for polypharmacy users — a majority of the T2D population
- Gaps in accessibility and multi-medication context are likely ADA compliance considerations, not just UX issues
- Biggest risk: Diane represents the highest-need, highest-retention-value segment; losing her means losing clinical outcomes data the client needs for payer partnerships
Persona 3: Jordan
Who they are
- 31-year-old UX researcher; recently diagnosed with T2D; highly health-literate and data-curious
- Distinct because they are actively trying to understand their condition, not just manage it — adherence is a means to an end (pattern recognition) rather than the goal itself
Primary job-to-be-done
- "See how my medication, food, sleep, and stress actually connect so I can make smarter decisions — not just follow instructions."
- The job is triggered constantly; Jordan checks health data the way others check social media
Top 3 pain points
- Adherence data lives in a silo — can't connect it to CGM readings, sleep data, or activity without manual effort; severity: High
- Nudges feel patronizing — generic encouragement messages read as noise to someone who already has high intrinsic motivation; severity: Medium
- No API or data export — can't pull data into their own dashboards; severity: Medium
Top 3 desired gains
- Integration with Apple Health, Oura, or Levels CGM data
- Insight-level coaching ("Your adherence was 95% on weeks you logged 7+ hours of sleep") rather than motivational copy
- Raw data access or CSV export
One unexpected insight
Jordan's engagement actually decreases when the app tries too hard to motivate them. High-autonomy users interpret frequent encouragement as a signal that the product doesn't trust them — which erodes trust in return. A "quiet mode" that strips notifications down to bare minimums could improve retention for this segment specifically.
Product fit assessment
- Current feature meets the functional minimum for Jordan but doesn't match their mental model of what a health tool should do
- Data portability and integrations are table-stakes for this persona, not differentiators
- Biggest risk: Jordan has high referral potential among younger, newly diagnosed patients — if they dismiss the app as unsophisticated, the word-of-mouth impact is disproportionate
Persona Comparison
| Dimension | Marcus | Diane | Jordan |
|---|---|---|---|
| Primary JTBD | Stay adherent without work disruption | Coordinate complex multi-med routine | Understand patterns across health data |
| Biggest pain | Inflexible reminder/logging flow | Fragmented polypharmacy management | Data siloed from rest of health stack |
| Product fit | Moderate | Weak | Weak–Moderate |
| Priority | High | High | Medium |
Shared Patterns
- Forgiveness over perfection: All three personas disengage when the app frames missed doses as failures. A "recovery flow" rather than a missed-dose alert would serve all three.
- Trust through transparency: Diane and Jordan (and implicitly Marcus) respond better to honest, qualified language than to confident clinical-sounding copy.
- Context the app doesn't have: Each persona's hardest moments occur in contexts the app is blind to — meetings, medical appointments, biometric patterns. Contextual awareness is the through-line.
Research Gaps
| Question | Why It Matters | Recommended Method |
|---|---|---|
| What percentage of users are managing 3+ medications? | Determines how urgently polypharmacy support needs to be scoped | Analytics review + onboarding survey |
| Do high-dismissal users re-dose later, or skip entirely? | Shapes whether flexible rescheduling is a retention fix or a clinical risk | Session replay + follow |