ChatGPT Prompts for Growth Hackers: Funnel Optimization, Viral Loops, Experiments

Published: July 2026 | 8 min read

Growth hacking is the discipline of finding leverage — small inputs that produce disproportionate outputs. AI can accelerate experiment design, surface non-obvious funnel leaks, and generate test variations faster than any human team. But "help me grow our user base" is not a prompt. Structured prompts that define the funnel stage, metric, constraint, and hypothesis turn AI into a growth analyst that never sleeps.

Below are production-ready ChatGPT prompts for growth hackers, organized by workflow. These are adapted from Skillent's Marketing AI Prompt Library.

Funnel Analysis & Optimization

1. Full Funnel Diagnostic

Role: Growth analyst
Task: Diagnose the growth funnel for [product type] and identify the biggest leak.
Funnel data:
- Visitors: [X]/month
- Signups: [X] ([X]% conversion)
- Activation: [X] ([X]% of signups)
- Retention (Day 30): [X]% 
- Revenue conversion: [X]% 
- Average revenue/user: $[X]
For each stage:
1. Is the conversion rate above or below industry benchmark? (provide estimate)
2. What's the biggest drop-off point?
3. 3 hypotheses for why users drop at this stage
4. 3 experiments to test each hypothesis (with expected impact)
5. Priority ranking by ICE score (Impact × Confidence × Ease)
Output: Funnel diagnostic with prioritized experiment backlog.
Format: Funnel diagram (text-based) + experiment table.

2. Activation Rate Optimizer

Role: Growth product manager
Task: Design 10 experiments to improve activation rate from [X]% to [target]%.
Product: [describe]
Current activation definition: [describe what counts as "activated"]
Current activation flow: [describe steps from signup to activation]
Experiment ideas across categories:
1. Onboarding changes (in-product)
2. Signup flow simplification
3. Email/onboarding sequence
4. In-app guidance/tooltips
5. Social proof placement
6. Pricing/offer adjustment
7. Feature gating changes
8. Timing changes (when to show what)
9. Personalization at first use
10. Notification/reminders
For each experiment:
- Hypothesis (if we [change], then [metric] will [direction] because [reasoning])
- Variable being changed
- Control and variant
- Success metric + secondary metric
- Estimated effort (hours)
- ICE score (Impact 1-10 × Confidence 1-10 × Ease 1-10)
Output: Prioritized experiment backlog ready for sprint planning.

Viral Growth Mechanics

3. Viral Loop Designer

Role: Viral growth strategist
Task: Design a viral loop for [product type].
Product: [describe — SaaS, consumer app, marketplace, etc.]
Current viral coefficient (K): [if known, or "unknown"]
Target audience: [describe]
Viral loop components to design:
1. Trigger (what motivates a user to invite)
   - Inherent (built into product use) vs. artificial (incentivized)
   - Emotional driver (utility, status, altruism, financial, FOMO)
2. Action (what the user does — invite, share, refer)
   - Friction level (1 click vs. multi-step)
   - Channel (email, social, link, in-person)
3. Reward (what the inviter gets)
   - Type (feature unlock, discount, credit, status)
   - Timing (instant vs. delayed)
4. Recipient experience (what happens when someone receives the invite)
   - Conversion path (how many steps to signup?)
   - First impression (what do they see?)
5. Re-engagement (how to bring the inviter back)
For each component:
- 3 options with pros/cons
- Recommendation with rationale
Calculate: Estimated K (viral coefficient) for recommended loop
Output: Full viral loop design with user flow diagram (text-based).
Include: Required product/engineering changes to implement.

4. Referral Program Architect

Role: Growth designer
Task: Design a referral program for [product].
Product: [describe]
Pricing model: [freemium / paid / subscription]
Target referral rate: [X]% of users referring
Program structure options (analyze 3):
1. Double-sided reward (both get [reward])
2. Single-sided referrer reward (referrer gets [reward])
3. Tiered/milestone rewards (refer 3 → [X], refer 10 → [Y])
For each option:
1. Reward type (cash, credit, feature, discount, physical)
2. Reward amount (calculate unit economics)
3. Eligibility rules (who can refer, minimum tenure)
4. Fraud prevention measures
5. Expected CPA through referrals vs. paid
6. Break-even analysis
Recommend: Which option maximizes K while protecting unit economics
Output: Referral program spec ready for engineering.
Include: Launch plan (how to announce, who to seed with, measurement framework).

Retention & Engagement

5. Retention Curve Analyzer

Role: Retention analyst
Task: Analyze this retention curve and recommend interventions.
Product: [describe]
Retention data (Day 1, 7, 30, 60, 90, 180, 365):
[X]%, [X]%, [X]%, [X]%, [X]%, [X]%, [X]%
Industry benchmark: [if known]
Analyze:
1. Retention curve shape (smiling = improving, flat = stable, frowning = decaying)
2. Where is the biggest drop? (Day 1? Week 1? Month 1?)
3. What cohort differences exist? (by signup source, persona, feature usage)
4. What does "retained" look like for our best users? (habits, frequency, features)
Interventions by timing:
- Day 0-1: [3 onboarding interventions]
- Day 1-7: [3 habit-formation interventions]
- Day 7-30: [3 value-realization interventions]
- Day 30+: [3 re-engagement interventions]
Output: Retention intervention plan with expected impact for each.
Include: Measurement plan — what to track, what defines success.

6. Cohort Analysis Framework

Role: Growth data analyst
Task: Design a cohort analysis framework for [product].
Product: [describe]
Key dimensions to segment by:
1. Acquisition source (organic, paid, referral, social)
2. Signup month/week (cohort)
3. User persona/type
4. Feature adoption at signup (which features used in first session)
5. Device/platform
Metrics to track per cohort:
1. D1/D7/D30 retention
2. Revenue per user (ARPU) at 30/60/90 days
3. Feature usage depth (how many features adopted)
4. Session frequency (times/week)
5. Churn rate (cumulative)
Output: Cohort analysis template with analysis questions to answer:
- Which acquisition source produces the highest LTV cohorts?
- Which features at signup predict Day 30 retention?
- Which cohorts are declining vs. improving?
Format: Analysis plan with specific SQL-style queries or spreadsheet formulas.

Experimentation

7. Growth Experiment Brief

Role: Growth experimentation lead
Task: Write an experiment brief for this growth test.
Hypothesis: [If we change X, then metric Y will increase because Z]
Experiment type: [A/B test / multivariate / before-after / holdout group]
Variable: [what's changing]
Control: [current state]
Variant(s): [new state(s)]
Primary metric: [the number that determines success]
Secondary metrics: [guardrail metrics — what we don't want to hurt]
Sample size: [calculate — include formula and assumptions]
Test duration: [X days — based on sample size and traffic]
Statistical significance threshold: [95% / 99%]
Segment: [all users / new users / specific cohort]
Risk assessment: [what could go wrong, blast radius]
Rollback plan: [how to revert if variant performs worse]
Output: One-page experiment brief ready for team review and engineering ticket.
Do NOT: Run tests without a pre-registered hypothesis. No peeking at results before the test concludes.

8. Experiment Prioritization Matrix

Role: Growth strategy lead
Task: Priorize these [X] growth experiment ideas using ICE scoring.
Experiments: [list all ideas with brief descriptions]
For each experiment, score:
1. Impact (1-10): If this works, how much does it move the primary metric?
2. Confidence (1-10): How sure are we this will work? (based on data, analogs, user research)
3. Ease (1-10): How easy is it to implement? (engineering hours, design needs, risk)
ICE Score = (Impact × Confidence × Ease) / 10
Sort experiments by ICE score (highest first).
For the top 5:
- Expected impact on primary metric
- Required resources (engineering, design, data)
- Dependencies (what needs to happen first)
- Timeline to results
Output: Prioritized experiment backlog for the next 2-3 sprints.
Format: Table with all scores + narrative for top picks.

Acquisition Channels

9. Channel Strategy Evaluator

Role: Growth channel strategist
Task: Evaluate acquisition channels for [product] and recommend a channel mix.
Product: [describe]
Target audience: [describe]
Pricing: [describe — LTV, CAC target]
Channels to evaluate:
1. SEO/content marketing
2. Paid search (Google Ads)
3. Paid social (Meta, LinkedIn, TikTok)
4. Organic social
5. Email marketing
6. Referral/word of mouth
7. Partnerships/integrations
8. Community/building
9. PR/media
10. Cold outbound (email/sales)
For each channel:
1. Expected CAC range
2. Expected volume (leads/month)
3. Time to results (weeks/months)
4. Required investment (time + money)
5. Scalability (can it 10x?)
6. Fit with audience (does our audience live here?)
Recommend: Top 3 channels to focus on for next 90 days
Output: Channel strategy matrix with phased rollout plan.
Include: What to stop doing (channels to pause or kill).

10. Pricing Experiment Designer

Role: Growth pricing strategist
Task: Design pricing experiments for [product].
Current pricing: [describe]
Current conversion rate: [X]%
LTV: $[X]
CAC: $[X]
Gross margin: [X]%
Experiments to design:
1. Price point test (same structure, different price)
2. Anchoring test (add premium tier to make current look reasonable)
3. Billing period test (monthly vs. annual discount)
4. Free trial length (7 vs. 14 vs. 30 days)
5. Freemium limits (what to gate vs. give free)
6. Bundle vs. standalone pricing
For each experiment:
- Hypothesis
- Variables
- Success metric (revenue per visitor, not just conversion rate)
- Risk (cannibalization, brand perception)
- Implementation complexity
Output: 6 pricing experiments with priority ranking.
Warning: Flag which experiments could permanently shift price perception (test carefully).

Analytics & Measurement

11. North Star Metric Definer

Role: Growth strategy lead
Task: Define the North Star Metric for [product] and its supporting metrics.
Product: [describe]
Business model: [SaaS / marketplace / ecommerce / media / marketplace]
Current primary metric: [describe, if any]
Analyze candidate metrics:
1. [Metric A] — pros, cons, when it leads vs. lags
2. [Metric B] — pros, cons, when it leads vs. lags
3. [Metric C] — pros, cons, when it leads vs. lags
For the recommended North Star:
1. Why it captures product value delivery
2. How it maps to revenue (lagging but predictive)
3. What inputs drive it (supporting metrics / input metrics)
4. How to measure it (data source, frequency, dashboard)
5. What it doesn't capture (blind spots)
6. Anti-patterns (how teams could game it)
Output: North Star framework with metric hierarchy:
North Star → Input Metrics → Supporting Metrics → Guardrail Metrics
Include: Dashboard mockup description (what charts to build).

12. Growth Review Report

Role: Growth analyst
Task: Write this month's growth review for [company].
Period: [month/year]
Key metrics:
- MRR/Revenue: $[X] (▲/▼ X% MoM)
- New users: [X] (▲/▼ X%)
- Activation rate: [X]% (▲/▼ Xpts)
- Retention (D30): [X]% (▲/▼ Xpts)
- NPS: [X] (▲/▼ X)
- CAC: $[X] (▲/▼ X%)
- LTV/CAC ratio: [X]
For each metric:
1. What happened ( factual)
2. Why it happened (correlation with experiments, changes, external events)
3. Is this expected? (vs. forecast/target)
Sections:
1. Executive summary (3 sentences — what matters this month)
2. Metric dashboard (table with MoM and YoY comparisons)
3. Experiment results (what we tested, what we learned)
4. Funnel analysis (where the biggest movement happened)
5. Next month's priorities (top 3 focus areas)
Output: 1-page growth review ready for leadership.
Tone: Data-first, no spin. Bad news before good news.

Best Practices for Growth Hacking AI Prompts

1. Always define the metric — "improve activation" means nothing without the specific number and definition

2. Include current benchmarks — AI needs baselines to identify what's broken

3. Ask for ICE scores — prioritization is half the battle in growth

4. Specify the funnel stage — acquisition, activation, retention, referral, or revenue

5. Always human-validate hypotheses — AI can generate ideas, but you need to verify they make sense for your product and users.

Related: Check out our guides on AI prompts for affiliate marketers and ChatGPT prompts for content marketers.