⚠️ Channel Adjustment for 2C Products
This skill defaults to dev / B2B channels.
2C Consumer Goods / Education / Apps: The main customer acquisition channels are
vertical communities + short videos + vertical KOLs. Select the top platform based on regional public data (e.g., short videos have surpassed Facebook in Indonesia/Thailand). Prioritize nano / micro vertical KOLs—the more followers, the lower the engagement rate; micro KOLs offer higher cost-effectiveness than mega KOLs. Complete 2C channel database + public sources →
gingiris-seo-geo/references/2c-adaptation.md
.
User Interview & Cold-Start Operations Practical Playbook
First-hand podcast cases and citation boundaries:
references/podcast-evidence.md
.
🌍 Language: Chinese | Interview Guide | Cold-Start Ops
Core Principles
"Founders of products will conduct intensive user interviews in the first six months. For example, HeyGen's founder did 937 interviews in six months; Wisperflow and Higgsfield both did over 500 interviews—continuous interviews and iterations." —— Ginger Iris
Key Insight: User interviews are not just research tools, but PMF discovery engines. Interview frequency is positively correlated with product success rate.
Token Top10 Interview Method (2026 Signature Method)
The shortest path to turn interviews from "research actions" into "daily operation cycles", fully reused in multiple AI Agent projects:
- Pull the Top10 user list from the backend based on token consumption / usage (refreshed weekly)
- Schedule shared-screen interviews, mandatory recording—observing real operations is better than listening to verbal descriptions
- Feed transcripts to AI, automatically extract feature requests / bugs and directly sync to Linear (or any issue tool)
- Follow up the next day: Notify users "Your request X has been fixed/scheduled"—feedback closure itself is a retention lever
- Rhythm: At least 2 interviews per day; after each interview, clip 5-10 minutes of key segments and send directly to product and R&D teams
Supporting Actions:
- New interviewers must first conduct mock interviews + full recording + highlight reviews before interviewing real users
- Two-in-one KOL interviews: Collaborate with YouTubers to conduct user interviews + usability testing simultaneously, getting two values from one collaboration; 20-30 minute video interviews offer the highest ROI
- Interview assetization: A paid user interview can simultaneously produce three outputs: functional requirements + SEO materials + reseller leads
Invitation Touchpoint Rhythm (2026 Update)
- Four touchpoints on Day 1/3/7/14: First send on Day 1 → follow up on Day 3 → Day 7 → Day 14
- Two channels: Email template + LinkedIn private message via email reverse lookup; if no reply to email, send a follow-up on LinkedIn
- Use broad reasons in the script ("The team wants to conduct user interviews in the early stage"), do not expose the real screening criteria to avoid users performing with presuppositions
Benchmark Anchors
Wisprflow: Manually onboarded the first 500 users and conducted one-on-one video interviews before launch, leveraging accumulated momentum for launch; HuggingFace: Conducted research + 500 user interviews in the first 8 months before scaling up.
2026-07 Supplement (First-hand Podcast Extraction):
- According to the former COO of AFFiNE: Gamma / Notion / HeyGen all conducted about 1,000 user interviews within six months, "truly solving the pain points mentioned by users".
- Common path for successful products like Gamma / HeyGen / Lovable: Polish product details through 500-1,000 interviews in six months before talking about growth—"Many people ignore the dirty and tiring work in the first six months to a year".
- Counterexample (anonymous project, reported): Skipping interview polishing and directly spending money on user acquisition, resulting in a single registration cost of tens of dollars and a single paid user cost of hundreds of dollars—"No logic, but deep pockets".
User Interview Execution Framework (5 Steps)
Step 1: Define Goals and Screen Users
User Priority:
| Priority | User Type | Value |
|---|
| P0 | Paid Users | Verified willingness to pay, highest value |
| P0 | High-frequency Active Users | Deep product understanding, most effective feedback |
| P1 | Competitor Users | Provide competitive perspectives |
| P1 | Churned Users | Expose real product problems |
| P2 | Registered Non-paid Users | Insights into conversion barriers |
Step 2: Invitation and Scheduling
Follow-up Rhythm: Day 1 → Day 3 → Day 7 → Day 10 (omnichannel: LinkedIn/Email/Telegram/Discord)
Core Invitation Script: Explain purpose + duration commitment (30 minutes) + recording explanation + value exchange
Step 3: Interview Execution (30-45 minutes)
Mandatory: 1 host + 1 note-taker + mandatory recording (including shared screen)
Core Question Framework:
- Background Understanding: Occupation, customer acquisition channel, usage duration, core features, competitors
- Workflow Exploration: What problem does it solve? How did you do it before? What changes have occurred now?
- Competitor Comparison: Which competitors have you used? What scenarios are they for? Our advantages and disadvantages?
- Pain Point Mining: What's unsatisfactory? Bugs? Hard-to-use features? "Magic wand" question
- Willingness to Pay: Have you paid for similar products before? How much are you willing to pay? What drives upgrades?
Key Technique: Observing screen sharing is better than listening to user descriptions. Let users demonstrate real operations and observe bottlenecks.
Step 4: Closing and Follow-up
- Invitation to ambassador program
- Request for referrals
- Complete interview review template within 24 hours
Step 5: Data Summary (Every 5-10 Interviews)
Summarize common needs/pain points → update feature priorities → revise user portraits → adjust product direction
Distinguishing Effective vs. Ineffective Interviews (2026-07 Supplement)
"An ineffective user interview is like giving an investor pitch—you talk 80% of the time, and the user is already distracted." —— Ginger Iris
| Dimension | Ineffective Interview | Effective Interview |
|---|
| Who Talks | You talk 80% of the time, user is distracted | Listen to users talk about how they use the product |
| What to Observe | Only listen to verbal descriptions, look at demo scripts | Let users share screens and observe real operations |
| Scenario | Hypothetical scenarios / you introduce the product | Real scenarios, real user cases |
| Mindset | Treat users as pure numbers | Put aside ego, make friends with users |
Judgment Basis: The user journey you constructed is completely different from the user's actual journey—only screen sharing can expose this gap.
Dual Logic of Willingness to Pay: "Saying willing to pay in interviews and actually paying are two different logics"—when it comes to real payment, everyone is extremely rational. Payment statements in interviews are only for directional reference; verification must be based on real payment behavior (corresponding to the "Four Timestamps for Payment Attribution" below).
Interview-driven Iteration Loop (Practical Path, 2026-07 Supplement)
- The first interview can last up to 2 hours: Deep users often understand scenarios, industries, and all competitors better than you, and can help you summarize the real value proposition and the current biggest comparative advantage.
- Sincere operations for in-depth feedback: An early overseas deep user later voluntarily wrote a 2,000-word long review with screenshots and GIFs, pointing out "spent too much time on functional scenarios users don't need" + specific bugs—"I wouldn't have spent two hours writing this if you weren't so sincere".
- Full-team dismantling → fixed within deadline: After receiving the long review, the whole team held a meeting to dismantle each item, fixing all bugs within 1-2 months.
- Closure notification: Follow up with users after fixing/scheduling "Your request X has been fixed"—feedback closure itself is a retention lever (echoing Step 4 of Token Top10)
- Pitfall Warning: If interview feedback is still decided solely by the top leader after being uploaded, a lot of user input will be ignored—product, R&D, and operations must jointly set goals and digest the essence of interviews.
Trust Building Speed Benchmark (Self-reported): No matter how high the position is, keep chatting directly with users on the frontline every day. The user trust built in 2 months can exceed that of colleagues who don't get involved in over a year—users will take the initiative to open up and endorse you across departments. High-frequency direct frontline communication is the prerequisite for interview quality.
Beta Test Design Framework
Three-tier Test Goal Structure
| Tier | Dimension | Data Collection Method |
|---|
| Value Verification Tier | Whether core pain points are solved | Interviews |
| Experience Perception Tier | Speed, usability, effectiveness | Observation (screen recording/remote connection) |
| Gap Identification Tier | Missing features, experience bottlenecks | Structured questionnaires + Bug feedback |
Test Task Design Principles
- Tasks simulate complete workflows, not isolated feature points
- Task length triggers "waiting" scenarios (expose performance issues)
- Provide prompt examples to reduce cognitive burden
- Clearly state "Beta instability is normal"
Core of Real-time Remote Testing
- Emphasize "observation" rather than "testing users"
- Think Aloud Protocol (talk while doing)
- Use loading waiting time to collect user background (underrated technique)
Five-question Method (Structured Closing)
- What kind of product do you think this is? (Let users label it)
- What is the most satisfying feature?
- Which link is the most laggy or confusing?
- If it launches officially tomorrow, would you be willing to pay? Why?
- What feature do you most want to see in the next version?
Cold-Start User Operations Methodology
Cold Touchpoint Funnel
Touch → Reply to invitation(~47%) → Deep tracking(~47%) → Qualified for testing(~24%) → Sustained activity
Key: 50 users matching ICP > 500 users with broad touchpoints
Messaging Iteration Method
| Message Type | Effect |
|---|
| Feature-driven (what it does) | Low conversion, attracts wrong user portraits |
| Scenario-driven (what problem it solves) | Conversion rate increases by ~83% |
| Pain point comparison (how you did it before) | Highest conversion |
Rule: The more specific the pain point description, the higher the conversion rate.
Two-tier Beta User System
| Tier | Portrait | Threshold | Benefits |
|---|
| Tier 1 Creative Partners | KOL/industry experts | Continuous usage >7 days + high-quality feedback | Pro access + direct founder contact + Case Study |
| Tier 2 Early Beta Users | Professional users matching ICP | Submit specific improvement suggestions | Early access + roadmap participation |
Activation and Payment Benchmarks (2026 Actual Measurement)
Interview Recruitment Priority and First-round Experiment
Priority order: Real paid/high-frequency users → Activated non-paid users → New registered non-activated users → Churned users → Broad target portraits. First send 10 personalized messages one by one, record replies, appointments, attendance, and effective insights, then decide whether to scale up. Prioritize letting users share screens to complete real tasks during interviews; opinions must be cross-validated with behavior logs, activation, and payment timestamps.
Activation must be defined by behavioral thresholds, not "registered / logged in":
| Product Type | Behavioral Threshold Examples |
|---|
| Agent Category | Complete ≥2 tasks |
| Conversation Category | ≥20 rounds of dialogue per day for 3 consecutive days |
| Tool Category | Start running tasks, or single usage exceeds 5 minutes |
| DAU Definition | Counted by unique message send, excluding internal users |
Conversion Qualification Line:
- Registration → Payment >5% is the qualification line: Pro-C 5-8%, SMB 10-15% (large C category 1.5-3%)
- Stage goals should be "achievable": Products with 0.1% payment rate should first set 1%/3 months, not directly 5%
Churn Operationalization: No follow-up within 24h after two conversations = churn, automatically trigger recovery emails—once churn is defined by behavior, recovery can be automated.
Four Timestamps for Payment Attribution: Registration time × Activation time × Payment time × Last login, cross-verify the four points to determine the real payment driver (which feature, interview, or touchpoint led to payment).
User Grading Standards
| Grade | Characteristics | Operation Strategy |
|---|
| S-level | High-frequency + paid + willing to promote | Priority for ambassador program, direct CEO contact |
| A-level | High-frequency + potential to pay | Priority for internal testing, regular follow-ups |
| B-level | Occasional usage + no clear payment intention | Guide to core features, drive conversion |
| C-level | Low activity + no payment intention | Reduce investment, focus on natural retention |
Churn Analysis Framework
| Churn Type | Typical Reasons |
|---|
| Performance Type | Device overheating/lag, loading timeout, crash |
| Workflow Disruption Type | Lack of downstream integration, insufficient feature completeness |
| ICP Mismatch Type | No data sovereignty, project-based with no long-term needs |
| Messaging Mismatch Type | Touchpoint value proposition ≠ actual product value |
Three Questions for Churned User Interviews:
- When did you stop using it? What were you doing at that time?
- Did you encounter any problems, or find an alternative?
- If we solve problem X, would you come back?
PMF Value Evaluation Matrix
Evaluate based on "Material Sovereignty × Usage Frequency":
| User Type | Material Sovereignty | PMF Value | Churn Risk |
|---|
| Independent Creators | Full sovereignty | ⭐⭐⭐⭐⭐ | Low |
| Institutional Decision-makers | Full sovereignty | ⭐⭐⭐⭐ | Price-sensitive |
| Freelancers | Temporary sovereignty | ⭐⭐⭐ | Churn after project ends |
| Institutional Executors | No sovereignty | ⭐⭐ | Concerns about data authorization |
Toolchain
| Category | Tool | Purpose |
|---|
| Questionnaire | Typeform | User research questionnaires |
| Recording | Grain, Fathom | Interview recording transcription |
| Database | Airtable | User database management |
| Scheduling | Calendly | Interview appointment |
| Touchpoint | LinkedIn, Discord | User contact and community |
Complete Reference Documents
| Document | Content |
|---|
| references/interview-guide.md | Complete user interview guide (HeyGen's 937-interview methodology, question framework, templates, FAQ) |
| references/cold-start-ops.md | Beta test design framework, cold-start user operations, churn analysis, reusable templates |
| references/templates.md | Interview review template, user portrait form, Feature Request Tracker |