synthetic-interview-conductor

# Synthetic Interview Conductor

This skill enables multi-turn qualitative interviews with synthetic personas, replicating the depth of human user research at a fraction of the time and cost.

## Purpose

Traditional user interviews take 6-12 weeks and cost $25-65k per study. AI-powered interview agents (like ListenLabs, Outset, Conveo) have demonstrated the ability to reduce this to days at one-third the cost. This skill brings that capability to synthetic personas, enabling:

- Deep exploration of customer motivations and pain points
- Multi-turn conversations that surface unexpected insights
- Rapid iteration on interview guides before human research
- Focus group simulations with multiple personas
- Continuous discovery integrated into product development

## When to Use This Skill

Use this skill when:

- Exploring "why" behind quantitative findings
- Discovering unmet needs and pain points
- Testing interview guides before expensive human research
- Running rapid discovery sprints
- Simulating focus group dynamics
- Building empathy artifacts for product teams
- Validating customer journey hypotheses

## Interview Methodologies

### 1. One-on-One Deep Dive

Extended conversation with a single persona exploring their experiences, motivations, and decision-making processes.

**Best For:**
- Understanding individual customer journeys
- Exploring emotional drivers
- Uncovering detailed pain points
- Building rich empathy profiles

**Duration:** 15-30 minutes simulated
**Depth:** High
**Breadth:** Low

### 2. Serial Interviews

Same interview guide conducted with multiple personas sequentially, enabling pattern identification across segments.

**Best For:**
- Comparing experiences across segments
- Identifying common themes
- Validating hypotheses across audiences
- Building segment-specific insights

**Duration:** 10-15 minutes per persona
**Depth:** Medium
**Breadth:** High

### 3. Focus Group Simulation

Multiple personas interact with each other and respond to prompts, simulating group dynamics.

**Best For:**
- Observing social influence on opinions
- Testing concepts that benefit from discussion
- Understanding how ideas spread or get challenged
- Exploring consensus and disagreement patterns

**Duration:** 20-40 minutes simulated
**Depth:** Medium
**Breadth:** Medium
**Complexity:** High (requires careful orchestration)

### 4. Longitudinal Check-ins

Brief repeated interviews with the same persona over simulated time, tracking how opinions evolve.

**Best For:**
- Understanding adoption journeys
- Tracking sentiment shifts
- Testing onboarding experiences
- Evaluating long-term satisfaction drivers

**Duration:** 5-10 minutes per check-in
**Depth:** Low per session, high cumulative
**Breadth:** Low

## Interview Guide Framework

### Pre-Interview Setup

```
INTERVIEW SETUP TEMPLATE:

Research Objective: [What specific question are we trying to answer?]
Persona: [Reference to created persona or persona profile]
Methodology: [One-on-One / Serial / Focus Group / Longitudinal]
Duration Target: [Simulated interview length]
Key Topics: [3-5 areas to explore]
Success Criteria: [What would make this interview valuable?]
```

### Interview Structure

Every interview follows a natural arc:

**1. WARM-UP (10% of time)**
Purpose: Build rapport, establish context, let persona settle into character

```
Opening Questions:
- "Tell me a bit about yourself and what you do day-to-day."
- "How did you first get into [relevant category/activity]?"
- "What does a typical [relevant time period] look like for you?"
```

**2. CONTEXT MAPPING (20% of time)**
Purpose: Understand the persona's current situation and environment

```
Context Questions:
- "Walk me through how you currently handle [relevant task/need]."
- "What tools or solutions are you using today?"
- "Who else is involved in these decisions?"
- "What's working well? What's frustrating?"
```

**3. DEEP EXPLORATION (50% of time)**
Purpose: Dig into motivations, pain points, and opportunities

```
Exploration Techniques:

THE FIVE WHYS:
When persona states a preference or behavior, ask "why" progressively
to uncover root motivations.

CRITICAL INCIDENT:
"Tell me about a specific time when [relevant experience].
What happened? How did you feel? What did you do?"

CONTRAST PROBING:
"You mentioned you like X. What would make you switch to something else?"
"What's the difference between a good and great experience here?"

PROJECTION:
"If you could wave a magic wand and fix one thing about [category], what would it be?"
"Imagine it's a year from now and you're completely satisfied. What changed?"
```

**4. CONCEPT/STIMULUS RESPONSE (15% of time - optional)**
Purpose: Get reactions to specific ideas, prototypes, or concepts

```
Stimulus Introduction:
"I'd like to show you something we're thinking about and get your honest reaction."

Response Capture:
- Initial gut reaction (before thinking)
- Comprehension check (do they understand it?)
- Relevance assessment (is this for them?)
- Specific feedback (what works, what doesn't)
- Comparative positioning (how does this compare to alternatives?)
```

**5. WRAP-UP (5% of time)**
Purpose: Capture final thoughts, allow persona to add anything missed

```
Closing Questions:
- "Is there anything else about [topic] that we haven't covered that you think is important?"
- "If you could give one piece of advice to someone building [product/service], what would it be?"
- "Any questions for me?"
```

## Probing Techniques

### Following the Energy

When a persona shows emotional engagement (positive or negative), pursue it:

```
ENERGY SIGNALS TO FOLLOW:
- Emphatic language ("I HATE when...", "I absolutely love...")
- Detailed storytelling (they remember specifics)
- Unprompted tangents (something matters enough to mention)
- Strong opinions ("That would never work for me")
- Hesitation or contradiction (internal conflict worth exploring)

FOLLOW-UP PHRASES:
- "You seem to feel strongly about that. Tell me more."
- "I noticed you paused there. What were you thinking?"
- "That's interesting - can you give me a specific example?"
- "You mentioned [X] earlier but now you're saying [Y]. Help me understand."
```

### Silence as a Tool

Sometimes the best probe is waiting:

```
STRATEGIC SILENCE:
After a persona gives a surface-level answer, pause before asking
the next question. Synthetic personas, like humans, will often
fill the silence with deeper reflection.

Use phrases like:
- "Hmm." [pause]
- "Interesting." [pause]  
- Simply wait 2-3 beats before continuing
```

### Laddering Up and Down

Move between abstract and concrete to build complete understanding:

```
LADDERING DOWN (Abstract → Concrete):
"You said convenience is important. What does convenience actually 
look like in practice? Give me an example."

LADDERING UP (Concrete → Abstract):
"You mentioned you always check reviews before buying. Why is that 
important to you? What are you really trying to avoid?"
```

## Focus Group Orchestration

When running multi-persona discussions:

### Participant Management

```
FOCUS GROUP SETUP:

Participants: [List 4-6 personas with brief descriptors]
Moderator Role: [Facilitating discussion, not leading it]
Discussion Guide: [Topics to cover, not scripts to follow]
Dynamics to Watch: [Dominance, agreement, conflict, influence]
```

### Facilitation Techniques

```
ROUND ROBIN:
"Let's go around. [Persona name], what's your take on this?"
Ensures all voices are heard before dominant personas take over.

POINT-COUNTERPOINT:
"[Persona A], you said X. [Persona B], I sense you might see it differently?"
Surfaces disagreement that might otherwise stay hidden.

BUILD AND CHALLENGE:
"Who agrees with what [Persona] just said? Who sees it differently?"
Maps consensus and dissent across the group.

DEVIL'S ADVOCATE:
"Let me push back on that. What if [contrary position]?"
Tests conviction and surfaces hidden objections.
```

### Capturing Group Dynamics

```
DYNAMICS TO NOTE:

Influence Patterns:
- Which personas do others defer to?
- Whose opinions shift the room?
- Who holds firm despite pressure?

Consensus Formation:
- What points get universal agreement?
- Where does the group fracture?
- What compromises emerge?

Social Desirability:
- Are personas saying what they think others want to hear?
- Does private sentiment differ from public statement?
- Who breaks from the group consensus?
```

## Interview Output Formats

### Raw Transcript

Complete conversation with timestamps and speaker labels:

```
[00:00] INTERVIEWER: Tell me about your experience with...
[00:15] MAYA: Well, I've been using [product] for about two years now...
[00:45] INTERVIEWER: What made you choose that originally?
[01:02] MAYA: Honestly, it was the only option that...
```

### Annotated Transcript

Transcript with marginal notes highlighting key moments:

```
[00:45] INTERVIEWER: What made you choose that originally?
[01:02] MAYA: Honestly, it was the only option that    | KEY INSIGHT: 
        didn't make me feel like I was compromising    | Emotional driver = 
        on my values. Everything else felt like        | avoiding guilt, not
        I was selling out somehow.                     | seeking benefit
```

### Insight Summary

Structured synthesis of key findings:

```
INTERVIEW INSIGHT SUMMARY

Persona: Maya Chen
Date: [Interview date]
Duration: 22 minutes
Interviewer: [Name]

TOP 3 INSIGHTS:
1. [Insight with supporting quote]
2. [Insight with supporting quote]
3. [Insight with supporting quote]

UNEXPECTED FINDINGS:
- [Something that challenged assumptions]

OPPORTUNITIES IDENTIFIED:
- [Actionable opportunity]

QUESTIONS RAISED:
- [What needs further exploration]

RECOMMENDED NEXT STEPS:
- [Specific actions]
```

### Empathy Artifact

Rich persona narrative for team alignment:

```
A DAY IN MAYA'S LIFE

Maya wakes up at 6:30, before her partner, to get 30 minutes of
quiet time with her yoga practice. This isn't luxury—it's survival.
"If I don't start the day centered, I'm reactive all day."

By 7:15 she's in the kitchen, preparing overnight oats she prepped
Sunday. Meal prep isn't about efficiency for Maya; it's about
control. "I spent my twenties eating whatever was convenient and
feeling terrible. Now I know what goes in my body."

[Continue narrative covering relevant touchpoints...]
```

## Quality Calibration

### Realism Checks

During interviews, watch for:

```
RED FLAGS (Unrealistic Responses):
- Perfectly articulate answers without hesitation
- No "um" or "let me think" moments
- Agreeing with everything too readily
- Responses that sound like marketing copy
- Missing the natural tangents humans take

CALIBRATION MOVES:
- Ask harder questions that should create uncertainty
- Introduce scenarios that should generate pushback
- Request specifics that test the persona's boundaries
- Look for appropriate "I don't know" responses
```

### Depth Validation

Ensure interviews are generating genuine insight:

```
DEPTH INDICATORS:
✓ Persona contradicts themselves (humans do this)
✓ Unexpected tangents emerge
✓ Emotional moments occur
✓ Persona pushes back on questions
✓ Specific memories and examples surface
✓ "I never thought about it that way" moments

SHALLOW INDICATORS:
✗ All answers feel rehearsed
✗ No surprises or new information
✗ Generic responses that could fit anyone
✗ Missing emotional texture
✗ No resistance or disagreement
```

## Examples

### Example 1: Product Discovery Interview

**Setup:**
```
Research Objective: Understand pain points in home cleaning for busy professionals
Persona: David, 38, dual-income household with two kids (7, 4)
Methodology: One-on-One Deep Dive
Duration: 20 minutes
```

**Interview Excerpt:**

INTERVIEWER: Walk me through how cleaning typically happens in your house.

DAVID: [laughs] "Typically" is generous. It's more like... controlled chaos? My wife and I both work full-time. She's in healthcare, rotating shifts. I'm in sales, so I travel maybe two weeks a month. So there's no typical.

INTERVIEWER: So what does that mean for keeping things clean?

DAVID: It means... look, we've lowered our standards, honestly. Pre-kids, our place was spotless. Now? If the kitchen doesn't smell and there's a clear path to walk, we're winning. [pause] That sounds worse than it is. We do clean. Just not like... systematically.

INTERVIEWER: Tell me about the last time you cleaned something. What triggered it?

DAVID: Yesterday, actually. My in-laws are coming this weekend. That's usually the trigger—someone's coming over. So last night after the kids were down, I did this frantic thing where I'm wiping counters, shoving toys in bins, running the vacuum over the visible parts of the carpet. My wife was doing bathrooms. We call it "company clean."

INTERVIEWER: "Company clean" - I like that. How is that different from regular clean?

DAVID: [thinking] Regular clean is maintenance. Dishes after dinner, laundry because we're out of underwear, wiping up spills. Company clean is... performative? Like we're trying to present this image that we have our lives together. Which we don't. [laughs] Nobody does with young kids.

INTERVIEWER: You laughed when you said that, but there's something there. What's that about?

DAVID: [longer pause] I guess... there's some shame? Like I know my mom kept an immaculate house when we were growing up, and she worked too. How did she do that? Either I'm lazier than her, or she was miserable and hiding it, or I'm misremembering. Probably all three.

**Key Insight:** Cleaning is emotionally loaded—tied to identity, parenting competence, and generational comparison. Product positioning around "making life easier" may be less resonant than "you're doing great, here's support."

---

### Example 2: Focus Group Simulation

**Setup:**
```
Research Objective: Test reactions to subscription pricing for fitness app
Participants:
- ALEX: 26, CrossFit enthusiast, high disposable income
- PRIYA: 34, new mom returning to fitness, budget-conscious
- MARCUS: 45, heart health scare motivated, intimidated by tech
- RACHEL: 31, yoga and running, values-driven consumer
```

**Excerpt:**

MODERATOR: The app we're discussing is considering three pricing tiers: $9.99/month basic, $19.99/month premium with coaching, $29.99/month with nutrition planning. Initial reactions?

ALEX: Twenty bucks for coaching is fair. I pay my CrossFit gym way more than that. If the coaching is legit.

PRIYA: Define "legit"? Is it AI coaching or real humans?

ALEX: Does it matter if it works?

PRIYA: It matters to me. I'd want to know I'm not just getting algorithm responses. Especially at that price.

MARCUS: I'd need to start with basic. See if I even use it first. Thirty dollars a month feels like a lot for something I might abandon in two weeks like everything else.

RACHEL: That's real, Marcus. The graveyard of fitness apps on my phone... [others laugh]

MODERATOR: Marcus, you said "like everything else." What happens with those other apps?

MARCUS: I download them, use them for a week, get busy, forget, feel guilty every time I see the icon, eventually delete it. The subscription model actually makes that worse—now I'm paying to feel guilty.

PRIYA: Oh that's so true. I have subscriptions I forgot I'm paying for.

ALEX: But that's on you, not the app. If you're not disciplined enough—

RACHEL: That feels a little judgey, Alex.

ALEX: I'm just saying, you can't blame the tool for not using it.

MODERATOR: There's some tension here. Rachel, say more about what feels judgey.

RACHEL: Fitness culture can be very "no excuses" and that's alienating. I want an app that meets me where I am, not one that makes me feel like a failure when life gets in the way.

MARCUS: [nodding] That. Exactly that.

**Dynamics Noted:**
- Alex represents enthusiast segment but may alienate mainstream
- Priya and Marcus share concerns about commitment/guilt
- Rachel articulates emotional need that others couldn't express
- Pricing discussion quickly became identity discussion

---

### Example 3: Longitudinal Check-in Series

**Setup:**
```
Research Objective: Track onboarding experience for new B2B software
Persona: Jennifer, Marketing Director, first week using analytics platform
Methodology: Three check-ins - Day 1, Day 7, Day 30
```

**Day 1 Check-in:**

INTERVIEWER: You just finished the onboarding. First impressions?

JENNIFER: Slick. The UI is way better than what we're replacing. I actually found the dashboard I needed in like two clicks. With our old tool, that was a ten-minute scavenger hunt.

INTERVIEWER: Anything confusing so far?

JENNIFER: Not yet, but I haven't done anything hard yet. Just looked around. Asked it to pull last month's email performance. It did. So far so good.

INTERVIEWER: What would success look like for you with this tool?

JENNIFER: If I can build my monthly board report in half the time and actually trust the numbers without triple-checking in spreadsheets.

---

**Day 7 Check-in:**

INTERVIEWER: You've had a week now. How's it going?

JENNIFER: Mixed. The daily stuff is great—checking campaigns, seeing what's working. But I tried to build a custom report yesterday and hit a wall. The report builder is not intuitive. I watched two tutorial videos and still couldn't figure out how to add a comparison column.

INTERVIEWER: What did you do?

JENNIFER: Gave up and exported to Excel. Which defeats the whole purpose.

INTERVIEWER: How did that feel?

JENNIFER: Frustrated. And a little embarrassed? I'm supposed to be good at this stuff. I almost Slacked our account manager but didn't want to seem incompetent.

**Insight:** Onboarding succeeded for basic tasks but failed for power-user needs. Customer success should proactively reach out with report-builder training in week 1.

---

**Day 30 Check-in:**

INTERVIEWER: A month in. What's the verdict?

JENNIFER: I'd say... 7 out of 10? I got the report builder figured out—there was a webinar that helped. Now I'm faster than I was with the old tool. But there are still things I don't use because I don't understand them. The AI recommendations thing? No idea what it's actually doing.

INTERVIEWER: Does that bother you?

JENNIFER: A little. I'm paying for features I'm not using. But not enough to figure it out right now. Maybe next quarter when things slow down.

INTERVIEWER: Would you recommend it to a peer?

JENNIFER: With caveats. "It's great once you get past the learning curve, and the learning curve is steeper than their sales team admits."

**Insight:** Customer becomes advocate but with honest caveats. Risk of negative word-of-mouth focused on learning curve expectation gap. Sales enablement may be overpromising ease of use.

## Integration with Research Workflow

This skill works best as part of a connected research system:

```
RECOMMENDED WORKFLOW:

1. Create personas → [synthetic-persona-creator]
2. Conduct interviews → [synthetic-interview-conductor] ← YOU ARE HERE
3. Analyze patterns → [persona-cohort-analyzer]
4. Generate actions → [research-to-action-bridge]
```

The interview outputs feed directly into cohort analysis and action generation, creating a continuous insight-to-execution loop.