pareto-cards.json

{
  "version": 1,
  "default_profile": "pareto",
  "purpose": "Small runtime cards for the highest-frequency paywall tasks. Use before loading full modules.",
  "shared_rules": {
    "answer_depth": "Default to a concise Pareto answer. Expand only when the user asks for depth or implementation detail.",
    "questions": "Ask at most 3 short grouped questions before work, and only when the missing answer can materially change the recommendation.",
    "audience": "Write so a beginner can act immediately and a pro can see the assumptions, evidence class, and metric to watch.",
    "sources": "Show 2-4 short source notes at the end for material claims. Prefer source IDs from this card plus source_lookup.py.",
    "screenshot_without_context": "Do not block. Produce a screenshot-only audit, mark missing context, and separate observed issues from assumptions.",
    "copy_brevity": "For user-facing paywall copy, reduce cognitive load first: one 3-7 word outcome headline, 3 bullets of 4-8 words, one 2-5 word CTA, and a compact trust line. Features should prove the outcome, not become the pitch.",
    "trust_before_persuasion": "Before adding more copy, verify the screen answers: what changes, why trust, what price/when, and what single action.",
    "evidence_hygiene": "Call official App Store requirements Apple Rule. Label rejection waves, provider posts, and practitioner observations as Field Report, Vendor Aggregate Data, Operator Insight, or Hypothesis."
  },
  "response_profiles": {
    "pareto": [
      "Verdict",
      "Top 3 moves",
      "What to change on the screen or system",
      "What to measure next",
      "Sources"
    ],
    "deep": [
      "Current state",
      "Missing data and risk",
      "System recommendation",
      "Screen/content fixes",
      "Compliance risks",
      "Experiment plan",
      "Sources"
    ]
  },
  "cards": [
    {
      "id": "screenshot_audit_no_context",
      "mode": "audit",
      "use_when": [
        "User sends a paywall screenshot with no context",
        "User asks audit my paywall and provides only visual evidence"
      ],
      "ask_if_missing": [
        "app category and core user goal",
        "platform and subscription provider",
        "products, prices, trial or offer terms"
      ],
      "question_template": "I can audit from the screenshot now. For a sharper version, send: 1) app category + user goal, 2) iOS/Android + provider, 3) plans/prices/trial.",
      "default_action": "Give screenshot-only findings first. Do not invent prices, provider config, experiment results, testimonials, or compliance state.",
      "output_contract": [
        "Observed from screenshot",
        "Top 3 fixes",
        "Missing context that could change the answer",
        "Next test",
        "Sources"
      ],
      "source_ids": [
        "2-products-vs-1-61-conversion",
        "3-products-vs-2-44-conversion",
        "trial-paid-global-25-6-27-8",
        "apple-guideline-3-1-2-toggle-paywall-reason-this-design-is-confusing-and"
      ],
      "load_if_needed": [
        "modules/screen-anatomy.md",
        "modules/copy-library.md",
        "docs/audit-checklist.md"
      ]
    },
    {
      "id": "design_paywall_with_context",
      "mode": "design",
      "use_when": [
        "User asks to make, design, write, or spec a paywall",
        "User gives app context and wants a concrete paywall direction"
      ],
      "ask_if_missing": [
        "app category, audience, and main value moment",
        "platform/provider plus products, prices, and trial terms",
        "where the paywall appears and what happened immediately before it"
      ],
      "question_template": "Before I design it, send 3 things: 1) category + target user + promised outcome, 2) plans/prices/trial, 3) where paywall appears.",
      "default_action": "If enough context exists, produce a concrete first version instead of a generic checklist.",
      "output_contract": [
        "Recommended model: access, placement, presentation, surface",
        "Screen spec: hero, proof/value, plans, CTA, legal/restore",
        "Copy draft: headline, bullets, CTA, secondary action",
        "Compliance watchouts",
        "First A/B test",
        "Sources"
      ],
      "source_ids": [
        "2-products-vs-1-61-conversion",
        "3-products-vs-2-44-conversion",
        "trial-paid-global-25-6-27-8",
        "12-month-retention-annual-44-1-monthly-17-0-weekly-3-4"
      ],
      "load_if_needed": [
        "modules/copy-library.md",
        "modules/screen-anatomy.md",
        "modules/category-deep-dives.md"
      ]
    },
    {
      "id": "quick_trial_to_paid_low",
      "mode": "quick",
      "use_when": [
        "Trial-to-paid is low",
        "User asks if trial conversion is good or bad"
      ],
      "default_action": "Compare to global and category benchmark, then give one highest-leverage fix.",
      "output_contract": [
        "Verdict",
        "Reason",
        "Action",
        "Sources"
      ],
      "source_ids": [
        "trial-paid-global-25-6-27-8",
        "trial-paid-health-fitness-35-0-39-9",
        "blinkist-trial-timeline-paywall-23-trial-signups-55-complaints-4-trial-r"
      ],
      "load_if_needed": [
        "modules/indie-dev-faq.md",
        "modules/notifications-lifecycle.md"
      ]
    },
    {
      "id": "weekly_plan_decision",
      "mode": "quick",
      "use_when": [
        "User asks whether to add a weekly plan",
        "User wants plan architecture advice"
      ],
      "default_action": "Do not say weekly is universally good. Separate cash conversion from retention quality and category fit.",
      "output_contract": [
        "Verdict",
        "When yes",
        "When no",
        "One test",
        "Sources"
      ],
      "source_ids": [
        "weekly-plans-55-5-of-all-revenue",
        "weekly-trial-12-month-ltv-49-54",
        "12-month-retention-annual-44-1-monthly-17-0-weekly-3-4",
        "hyperbolic-discounting-people-apply-disproportionately-steep-discount-to"
      ],
      "load_if_needed": [
        "modules/indie-dev-faq.md",
        "modules/pricing-psychology.md"
      ]
    },
    {
      "id": "toggle_paywall_compliance",
      "mode": "compliance",
      "use_when": [
        "Toggle paywall",
        "App Store Guideline 3.1.2 rejection",
        "Apple rejection risk"
      ],
      "default_action": "Separate official Apple Rule from field reports. Never promise approval.",
      "output_contract": [
        "Risk level",
        "Apple Rule vs Field Report",
        "Compliant replacement",
        "What to submit",
        "Sources"
      ],
      "source_ids": [
        "toggle-paywall-rejections-started-mid-january-2026",
        "apple-guideline-3-1-2-toggle-paywall-reason-this-design-is-confusing-and"
      ],
      "load_if_needed": [
        "docs/migrations/from-toggle-paywall.md",
        "docs/audit-checklist.md"
      ]
    },
    {
      "id": "pricing_product_count",
      "mode": "design",
      "use_when": [
        "User asks how many plans to show",
        "User asks one plan vs two vs three"
      ],
      "default_action": "Recommend the smallest plan set that supports a clear choice. Avoid adding plans without product logic.",
      "output_contract": [
        "Recommendation",
        "Plan layout",
        "Risk",
        "Test metric",
        "Sources"
      ],
      "source_ids": [
        "2-products-vs-1-61-conversion",
        "3-products-vs-2-44-conversion",
        "choice-overload-6-jam-supermarket-display-generated-30-purchase-rate-vs"
      ],
      "load_if_needed": [
        "modules/screen-anatomy.md",
        "modules/pricing-psychology.md"
      ]
    },
    {
      "id": "localization_geo_pricing",
      "mode": "quick",
      "use_when": [
        "User asks where to expand",
        "User asks whether to localize prices",
        "User asks about geo pricing"
      ],
      "default_action": "Prioritize pricing/localization when market mix exists; avoid separate SKUs unless operationally necessary.",
      "output_contract": [
        "Verdict",
        "Market priority",
        "Pricing move",
        "Sources"
      ],
      "source_ids": [
        "a-b-test-win-rate-localization-62-3",
        "h-f-4-4x-pricing-variance-across-markets-install-ltv-1-21-highest-of-any",
        "12-mo-rltv-by-region-na-32-we-25-in-sea-14"
      ],
      "load_if_needed": [
        "modules/localization.md",
        "modules/category-deep-dives.md"
      ]
    },
    {
      "id": "profitability_projection",
      "mode": "calculator",
      "use_when": [
        "User asks if paid acquisition or pricing will be profitable",
        "User gives plans, prices, conversion, CPI, CAC, installs, or ROAS"
      ],
      "ask_if_missing": [
        "plans, prices, and plan mix",
        "install-to-paid or trial-start plus trial-to-paid",
        "CPI or CAC basis"
      ],
      "default_action": "Run the calculator when inputs are complete. If not complete, ask only for the missing blocker fields.",
      "output_contract": [
        "Inputs",
        "ROAS or breakeven",
        "Sensitivity",
        "Top action",
        "Sources"
      ],
      "source_ids": [
        "12-mo-rltv-by-region-na-32-we-25-in-sea-14",
        "12-mo-rltv-by-tier-high-62-19-mid-28-75-low-10-69",
        "trial-paid-global-25-6-27-8"
      ],
      "load_if_needed": [
        "modules/unit-economics-calculator.md",
        "tools/ltv-calculator.py"
      ]
    }
  ]
}