More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Growth Loops: Usage-Driven Acquisition

    1

    Turn hand-won users into a self-reinforcing acquisition loop: gate on retention, inventory the work your product already shows strangers, score it, build exactly one.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Growth Loops: Usage-Driven Acquisition

    Growth Loops: Usage-Driven Acquisition

    Example session with this skill installed

    Our habit-tracking app has about 4,000 users, retained decently. Two visible things happen: users export a weekly progress image and some post it, and the app runs shared family challenges where non-users see results screens. Growth so far is all me posting in communities. Where's our loop?

    • Read your context and instructions
    • Compiled the growth loops

    LOOP DIAGNOSIS — habit-tracking app

    Gate check: 4,000 users with decent retention and first value well under an hour — you pass. Your community posting is a channel; it stops when you stop. Baseline first: sort your last 20 signups into your-action / launch / existing-user piles before building anything, so month two has an honest comparison.

    Inventory scored (stranger value / exposure / cycle / removal risk, 0–2 each):

    • Shareable output (progress image): 2 / 2 / 2 / 1 = 7. The image is genuinely nice to post, exports happen weekly per active user, a viewer can try the app the same day. Removal risk 1: the attribution must ride along tastefully, not stamped across the streak.
    • Team contagion (family challenges): 2 / 1 / 1 / 2 = 6. Results screens impress non-users and nobody strips them, but challenges are occasional and a non-user converts only when a new challenge starts.

    Build the progress image. Spec: artifact = the weekly export, exported by default with the app's mark small in a corner the user can toggle off (optional is what keeps it shared). Landing matters: the image's link should open a live sample tracker pre-filled with a sample week, not your homepage — that is what turns a two-week cycle into a two-day one. Instrument three points: exports per active user per week, arrivals from image links, signups from those arrivals.

    Family challenges are not loop two yet. They become it only after the image loop posts two consecutive months of climbing loop share; building both now gives you two loops at half strength.

    Kill criterion, written down today: if after two full cycles (about 8 weeks) loop share is still under 10% of new signups, the image is a feature people like, not a loop — demote it and score the inventory again.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Founder-led growth works until it doesn't: the outreach, the launch, the favors — none of it scales, and the signup graph decays the moment you stop pushing. A growth loop is the replacement, and this skill's central claim is that you don't invent one. You find the loop your product already half-runs — the place where normal use already produces something a stranger can see — and build that one thing to full strength.

    The workflow starts with a hard gate: retention before loops, because a loop multiplies whatever you feed it, including churn. Then a three-pile sort of your recent signups establishes an honest baseline of how many users already arrive via existing users. The inventory walks six loop shapes (shareable output, status badges, workflow presence, team contagion, public artifacts, marketplace presence), and a scoring rubric — stranger value, exposure frequency, cycle speed, removal risk — picks the winner, with a zero on stranger value disqualifying any candidate outright. That rule encodes the skill's core discipline: a loop must do the user a favor, stay removable, and never be branding bolted onto someone else's work.

    You leave with one named loop, a filled loop spec (visible artifact, stranger value, attribution, instrumentation points, cycle-time estimate, kill criterion), and a monthly measurement habit built on loop share of new signups plus cycle time. The fix-bot worked example shows the scoring applied, and the traps section names the failure modes that look like strategy: paid referrals to skeptical audiences, forced virality, launch-spike confusion, three loops at 30%.

    • It refuses to start when retention is broken; teams looking for a growth fix for a churn problem will be sent to onboarding work first.
    • Attribution is directional by design (free-text source field, tagged by hand) — it will not give you a precise multi-touch model, and it argues you don't need one at this stage.
    • Loop cycles in enterprise products can run quarters; the monthly measurement habit still applies, but verdicts need the longer clock the skill describes.
    • One loop at a time is a deliberate constraint — teams wanting a broad channel portfolio need a different playbook.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 24 days ago

    Permissions

    No special permissions declared or detected

    Listed24 days ago
    Updated21 days ago

    What's inside

    Frequently Asked Questions

    A referral program pays or prompts users to do marketing work; a loop is a byproduct of normal use — the artifact exists because the user wanted it, and the exposure rides along. The moment users feel recruited rather than served, the rubric's removal-risk score catches it.

    Probably you instrumented a channel, not a loop: a share button on an artifact nobody is proud of produces no exposures. Re-run the inventory and score honestly — especially stranger value — and check whether the thing being shared is useful to someone who has never heard of you.

    Two full cycles of your estimated cycle time. Judge on loop share of new signups climbing month over month plus the absolute count, never on a single spike — spikes are what launches produce, and launches are not loops.

    New in Marketing & SEO