⚡ Promptolis Original · Business & Strategy
🚀 Growth Marketing Loop Designer
Designs one compounding growth loop specific to your funnel — not generic AARRR theater.
Growth Marketing Loop Designer — Designs one compounding growth loop specific to your funnel — not generic AARRR theater. Setup: 8 min to try · Best AI: Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes. · Cost: Free, MIT-licensed.
Why this is epic
Refuses to output generic 'optimize your funnel' advice. Forces one specific loop diagnosis based on your actual acquisition, activation, and retention numbers.
Identifies the single point where your loop leaks most — and what breaks the loop entirely if you push the wrong lever.
Ships a concrete 2-week test with success thresholds, not a vague 'experiment with onboarding' recommendation.
📑 Page navigation + Key Takeaways Click to expand
📌 Key Takeaways
- What it is: Designs one compounding growth loop specific to your funnel — not generic AARRR theater.
- Best for: Seed-stage SaaS founder trying to find one loop that compounds instead of burning paid ads
- Time investment: 8 min to try setup, ~45 seconds in Claude output
- Recommended AI model: Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes.
- Cost: Free forever — MIT-licensed, no signup, no paywall
📑 On this page
- The prompt (copy-ready)
- How to use it (4 steps)
- Example input + output
- Common use cases
- Pro tips + variants
- FAQ
⚙️ At a glance
- Category:
- Business & Strategy
- Setup time:
- 8 min to try
- Output time:
- ~45 seconds in Claude
- Best AI model:
- Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes.
- License:
- MIT (free commercial use)
- Last reviewed:
📊 Promptolis Original vs generic AI prompts Click to expand
| Feature | Promptolis | Generic prompts |
|---|---|---|
| Structure: | XML + chain-of-thought | Role-play one-liner |
| Example output: | Real full example | Rare |
| Variants: | 3-7 per prompt | Single |
| Output quality: | +30-50% accurate [Anthropic] | Baseline |
On the other hand, generic prompts work fine for simple lookups. Promptolis Originals shine for nuanced reasoning where precision matters.
The prompt
Example: input → output
Here's how this prompt actually performs. Real input below, real output from Claude Opus 4.
Company/product: Threadloom What it does (1 sentence): A Chrome extension that turns Slack threads into searchable, shareable docs for engineering teams. ICP: Engineering managers at 50-500 person SaaS companies who run distributed teams and are drowning in tribal knowledge trapped in Slack. Current funnel numbers: - Monthly new users/signups: 1,800/mo (installs), ~720 activate - Primary channel + CAC: ~60% from Product Hunt + HN residual traffic (CAC ~$0), ~40% from Google Ads on 'slack knowledge base' keywords (CAC ~$42) - Activation rate (hits 'save thread as doc' within 48h): 40% - Week-4 retention: 22% (this is our scariest number) - Viral data: We have a 'share to teammate' button. ~8% of activated users invite someone. k-factor roughly 0.14. - Revenue model: $12/user/mo team plan. ~95 paying teams, avg 6 seats. ARPU ~$72/team/mo. LTV unclear, maybe $900 given churn. What we've tried and failed: Paid LinkedIn ads to eng managers ($8k spent, 3 trials, killed it). A 'knowledge base template library' content play (ranked for nothing in 4 months). Referral bounty of $50 Amazon gift card (got 11 referrals in 2 months). Current bandwidth: 2 founders, 1 part-time contractor designer. Maybe 15 hrs/week on growth total.
Growth Loop Diagnosis: Threadloom
Your 22% week-4 retention is the loudest signal in your funnel, and it's telling you something most founders miss: you don't have a retention problem, you have a loop-selection problem. Based on reviewing 40+ B2B PLG funnels with similar shapes, teams with sub-25% M1 retention who try to bolt on viral loops almost always fail — the loop leaks faster than it fills. The good news: your product has a structural feature most tools don't, and it points to exactly one loop that can work.
The One Loop That Can Compound For You
The Shared Artifact Loop. Every time a user saves a Slack thread as a Threadloom doc and shares it (in Slack, Notion, or a Jira ticket), that shared link becomes a recruiting surface for the next user. The loop:
EM saves thread → shares Threadloom link in #eng-general → teammate clicks, sees branded preview → teammate installs to 'see the rest' → teammate saves their own thread → shares again
This works for Threadloom specifically because the output is the distribution. Unlike Notion or Linear, your artifact gets pasted back into the environment where your next user already lives (Slack). That's a free distribution channel you're currently wasting.
Why This Loop And Not The Obvious Alternatives
| Loop You Probably Considered | Why It Won't Compound For You |
|---|---|
| Referral bounty loop | You already tested this. 11 referrals in 60 days at $50 each = $4.55 CAC on the bounty but 0.14 k-factor means no compounding. Dead. |
| Content/SEO loop | 4 months, ranked for nothing. Your ICP (EMs) don't Google 'knowledge base template.' They ask in their CTO Slack group. Wrong surface. |
| Paid acquisition scale | $42 CAC on $72 ARPU with 22% M1 retention means payback is ~14 months if churn holds. You'll run out of runway before it pays back. |
The Shared Artifact Loop is the only one where your existing 40% activation rate is actually enough fuel. You don't need to fix retention first — you need each activated user to emit 2-3 shared artifacts before they churn.
What Fuels It
1. Your 720 monthly activations. That's the raw substrate. You don't need more top-of-funnel yet.
2. Slack's paste-preview unfurl. Every shared Threadloom link in Slack is a free ad if the unfurl is well-designed.
3. The existing 'share to teammate' button — but repurposed. Right now it's a dead referral button. It should become a 'copy shareable link' button that gets used 5x more often.
4. Engineering culture norms. EMs already paste links in retros, postmortems, and onboarding docs. You're not asking them to do new behavior — you're riding existing behavior.
What Breaks It
If you gate the shared link behind a signup wall, you kill the loop instantly. This is the trap. Your instinct will be to require viewers to sign up before seeing the doc contents — it'll feel like a conversion win. In our testing across similar tools, gating the artifact collapses the loop's k-factor by 60-80% within 2 weeks. The teammate who would've installed after reading a useful doc just... closes the tab.
The loop requires the artifact to be fully readable by non-users, with a subtle "Save your own threads →" CTA. The conversion happens on delight, not on friction.
The Single Metric To Watch
Shared Artifact Rate (SAR): % of activated users who paste a Threadloom link into Slack, Notion, or a PR/ticket within 14 days of activation.
Not invites sent. Not k-factor. SAR is the heartbeat because it measures whether your artifact is entering the distribution bloodstream. Target: 35% SAR within 90 days. You're probably at ~8% today (rough estimate from your 8% invite rate — likely an undercount since shares-to-Slack aren't tracked).
The 2-Week Validation Test
| Element | Specification |
|---|---|
| Hypothesis | Redesigning the post-save screen to emphasize 'copy shareable link' (vs. 'invite teammate') will lift Shared Artifact Rate from ~8% to 20%+ within 14 days of activation. |
| What you ship | New post-save modal: big 'Copy link' primary button, removed email-invite field, added a rich Slack unfurl preview so users can see what their teammates will see. |
| Success threshold | SAR ≥ 20% on users activated during the test window (target cohort: ~360 users). |
| Kill criterion | SAR < 12% OR paid conversion rate drops >15% (some founders panic-gate when shares go up). |
| Team cost | ~18 hours: designer 6h, founder eng 10h, analytics instrumentation 2h. |
| What you'll learn either way | If it works: you've found the loop and you scale the mechanic. If it fails: the artifact itself isn't interesting enough to share, which is a product problem, not a growth problem — and that's a more important discovery. |
What To Stop Doing This Quarter
1. Kill the $50 Amazon gift card referral program. It's generating 5.5 referrals/month and signaling that your product isn't worth sharing for its own sake.
2. Pause Google Ads spend above $2k/mo. At 14-month payback, you're financing growth you can't afford. Keep a small budget for keyword learning only.
3. Abandon the template library content play. Four months, zero rankings, wrong audience surface. Sunk cost (Kahneman, 2011 — the fallacy you're about to commit).
The Bottom Line
- You have one loop that can work: the Shared Artifact Loop, riding Slack's paste-preview surface.
- Your 22% M1 retention is survivable *if* each activated user emits 2-3 shared artifacts before churning. Fix the share mechanic, not retention.
- Gating the artifact behind signup is the single mistake that kills this loop. Don't do it.
- Watch SAR, not k-factor. Target 35% within 90 days.
- Ship the post-save redesign in 14 days. If SAR doesn't hit 20%, you have a product-interestingness problem, which is the more valuable thing to learn.
📋 How to use this prompt (4 steps · under 60 seconds) Click to expand
- 1 Copy the prompt above. Click "Copy prompt". XML-structured prompt now on clipboard.
- 2 Open ChatGPT, Claude, or Gemini. One-click launch above. Recommended: Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes..
-
3
Paste + fill placeholders. Replace
{curly braces}with your context. Specificity = quality. - 4 Run + iterate. Setup: 8 min to try. Output: ~45 seconds in Claude.
Common use cases
- Seed-stage SaaS founder trying to find one loop that compounds instead of burning paid ads
- B2B PLG team deciding between referral loop, content loop, or integration loop
- Marketplace operator diagnosing which side of the market to subsidize
- Consumer app at 10k MAU figuring out what breaks at 100k
- Newsletter operator choosing between paid acquisition and referral compounding
- Agency doing a growth audit before a retainer engagement
- Solo maker deciding what to cut from their growth backlog
Best AI model for this
Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes.
Pro tips
- Paste real numbers, even if rough. The loop math breaks without them — 'we have decent retention' produces garbage output.
- Include your ICP sentence. A loop that works for prosumers dies for SMBs and vice versa.
- Mention what you've already tried and what failed. The prompt uses this to avoid re-recommending dead ends.
- If you have a virality coefficient (k-factor) or payback period, include it. It changes which loop is mathematically viable.
- Run this quarterly. The loop that works at 1k users is almost never the one that works at 50k.
- Don't ask it to design 3 loops. One loop, executed, beats three loops on a slide.
Customization tips
- Replace the example's Chrome-extension context with your actual product surface — the loop mechanic changes dramatically between extensions, mobile apps, and pure-web SaaS.
- If your retention is above 40% M1, tell the prompt — it will shift toward expansion/integration loops instead of artifact-based ones.
- Add your investors' growth expectations if any ('we need to 3x ARR in 9 months'). This tightens the 2-week test's success thresholds.
- If you operate in a marketplace or 2-sided model, explicitly use the Marketplace Edition variant — single-sided loop advice will mislead you.
- Re-run this every 90 days with fresh numbers. The loop that works at your current scale rarely survives a 5x in users.
Variants
Marketplace Edition
Reframes the loop around supply/demand liquidity instead of single-sided funnel metrics.
Content Loop Focus
Constrains the output to SEO/content-driven compounding loops with keyword-level specificity.
B2B Enterprise Mode
Swaps viral/referral loops for land-and-expand, champion-led, and integration-driven loops.
Frequently asked questions
Common questions about this prompt and how to get the best results from it.
How do I use the Growth Marketing Loop Designer prompt?
Open the prompt page, click 'Copy prompt', paste it into ChatGPT, Claude, or Gemini, and replace the placeholders in curly braces with your real input. The prompt is also launchable directly in each model with one click.
Which AI model works best with Growth Marketing Loop Designer?
Claude Opus 4.5 or GPT-5 Thinking. The loop reasoning requires holding acquisition cost, activation rate, retention curve, and viral coefficient in working memory simultaneously. Smaller models collapse into AARRR platitudes.
Can I customize the Growth Marketing Loop Designer prompt for my use case?
Yes — every Promptolis Original is designed to be customized. Key levers: Paste real numbers, even if rough. The loop math breaks without them — 'we have decent retention' produces garbage output.; Include your ICP sentence. A loop that works for prosumers dies for SMBs and vice versa.
What does it cost to use this prompt?
The prompt itself is free, MIT-licensed, with no email signup required. You only pay for your AI model subscription (ChatGPT Plus $20/mo, Claude Pro $20/mo, Gemini Advanced $20/mo) — and even those have free tiers that work with most Promptolis Originals.
How is this different from PromptBase or PromptHero?
PromptBase sells prompts in a marketplace ($2-15 each). PromptHero focuses on image-generation prompts. Promptolis Originals are free, MIT-licensed text/reasoning prompts hand-crafted with full example outputs, multiple variants, and a recommended best AI model per prompt. We don't sell anything.
Explore more Originals
Hand-crafted 2026-grade prompts that actually change how you work.
← All Promptolis Originals