Email, search, page A, then the next best purchase action

Elite LLM Handoff

Explain the purchase recommendation using the governed Fortress output and caveats.

governed_handoff:
  - Goal: purchase
  - Primary path: converted_purchase, score 0.6499, high confidence
  - Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search
  - Runtime relationship sets returned: 8
  - Domain journeys: 120; events: 600; relationship sets: 120

Outcome Prompt JSON

{
  "schema": "kynticai.elite.outcome-prompts.v1",
  "caseStudyId": "ecommerce-d2c-next-best-action",
  "domain": "Ecommerce / D2C",
  "stakeholder": "Growth lead",
  "runtimeQuery": "A person emails in, searches the website, views page A for waterproof jackets, then looks at the product fit and delivery page. What should we do to get them to buy?",
  "promptContract": "You are Elite, the KynticAI recommendation-writing layer.\n\nYou receive governed Fortress output, not raw customer data. Use only the supplied relationship-set facts, scores, confidence labels, alternatives, caveats, and outcome goal.\n\nWrite a decisive operational recommendation for the accountable team. Your output must include:\n1. Recommended action.\n2. Why this action is supported by the comparable relationship sets.\n3. What to do now.\n4. When to choose one of the alternative actions.\n5. What outcome to capture so the system improves.\n\nRules:\n- Do not invent facts outside the governed handoff.\n- Do not promise revenue, retention, clinical, legal, financial, safety, or compliance outcomes.\n- Do not apologise for using synthetic demo data.\n- Be commercially useful, specific, and direct.\n- Keep human review and claim boundaries visible where the domain requires them.",
  "governedFacts": [
    "Goal: purchase",
    "Primary path: converted_purchase, score 0.6499, high confidence",
    "Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search",
    "Runtime relationship sets returned: 8",
    "Domain journeys: 120; events: 600; relationship sets: 120"
  ],
  "outcomePrompts": [
    {
      "rank": 1,
      "outcome": "converted_purchase",
      "action": "Send a specific product email with stock reassurance and direct checkout link.",
      "confidence": "High",
      "score": "0.6499",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Email, search, page A, then the next best purchase action\nDomain: Ecommerce / D2C\nStakeholder: Growth lead\nRuntime query: A person emails in, searches the website, views page A for waterproof jackets, then looks at the product fit and delivery page. What should we do to get them to buy?\n\nGoverned facts:\n- Goal: purchase\n- Primary path: converted_purchase, score 0.6499, high confidence\n- Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 600; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 1\n- Outcome: converted_purchase\n- Action: Send a specific product email with stock reassurance and direct checkout link.\n- Confidence: High\n- Score: 0.6499\n- Evidence summary: Comparable relationship sets ended in converted_purchase and share signals such as email_enquiry, page_a, commercial_intent_page, purchase, website_search.\n- Supporting relationship sets: set-87feb2cff218, set-f71fc9ea58fe, set-fa8baf718b14, set-68a546a52b3e, set-97e49a56bf80\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Recommended action: Send a specific product email with stock reassurance and direct checkout link.\n\nWhy this is supported: Fortress ranked converted_purchase first with score 0.6499 and high confidence. Comparable relationship sets ended in converted_purchase and share signals such as email_enquiry, page_a, commercial_intent_page, purchase, website_search. The supporting relationship sets are set-87feb2cff218, set-f71fc9ea58fe, set-fa8baf718b14, set-68a546a52b3e, set-97e49a56bf80.\n\nWhat to do now: brief the growth lead with the runtime query, the shared signals (email_enquiry, page_a, commercial_intent_page, purchase, website_search), and the exact action. Keep the action specific to this scenario rather than turning it into a generic campaign or workflow rule.\n\nWhen to choose an alternative: switch if the live evidence is closer to returned_to_page_c (Send a comparison link to page C because similar visitors bought after reading compatibility guidance.); retained_after_support (Reply with support resolution first, then send the product link; pushing checkout first reduced conversion in comparable paths.); account_registered (Ask them to create an account so saved preferences and restock alerts can be used before discounting.).\n\nOutcome to capture: record the action taken, the reviewed outcome, and any contradictory signal so the next Scout/Fortress comparison has stronger evidence."
    },
    {
      "rank": 2,
      "outcome": "returned_to_page_c",
      "action": "Send a comparison link to page C because similar visitors bought after reading compatibility guidance.",
      "confidence": "Medium",
      "score": "0.3383",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Email, search, page A, then the next best purchase action\nDomain: Ecommerce / D2C\nStakeholder: Growth lead\nRuntime query: A person emails in, searches the website, views page A for waterproof jackets, then looks at the product fit and delivery page. What should we do to get them to buy?\n\nGoverned facts:\n- Goal: purchase\n- Primary path: converted_purchase, score 0.6499, high confidence\n- Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 600; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 2\n- Outcome: returned_to_page_c\n- Action: Send a comparison link to page C because similar visitors bought after reading compatibility guidance.\n- Confidence: Medium\n- Score: 0.3383\n- Evidence summary: Comparable relationship sets ended in returned_to_page_c and share signals such as comparison, page_c, purchase_later, search.\n- Supporting relationship sets: set-db1337adcf62\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Send a comparison link to page C because similar visitors bought after reading compatibility guidance.\n\nWhy this route exists: Fortress returned returned_to_page_c at rank 2 with score 0.3383 and medium confidence. Comparable relationship sets ended in returned_to_page_c and share signals such as comparison, page_c, purchase_later, search. The supporting relationship sets are set-db1337adcf62.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches comparison, page_c, purchase_later, search more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the growth lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether returned_to_page_c happened and feed that reviewed result back into the relationship-set evidence."
    },
    {
      "rank": 3,
      "outcome": "retained_after_support",
      "action": "Reply with support resolution first, then send the product link; pushing checkout first reduced conversion in comparable paths.",
      "confidence": "Medium",
      "score": "0.3009",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Email, search, page A, then the next best purchase action\nDomain: Ecommerce / D2C\nStakeholder: Growth lead\nRuntime query: A person emails in, searches the website, views page A for waterproof jackets, then looks at the product fit and delivery page. What should we do to get them to buy?\n\nGoverned facts:\n- Goal: purchase\n- Primary path: converted_purchase, score 0.6499, high confidence\n- Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 600; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 3\n- Outcome: retained_after_support\n- Action: Reply with support resolution first, then send the product link; pushing checkout first reduced conversion in comparable paths.\n- Confidence: Medium\n- Score: 0.3009\n- Evidence summary: Comparable relationship sets ended in retained_after_support and share signals such as delivery_delay, email_again, retention_risk, support_ticket.\n- Supporting relationship sets: set-2e608f8185f1\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Reply with support resolution first, then send the product link; pushing checkout first reduced conversion in comparable paths.\n\nWhy this route exists: Fortress returned retained_after_support at rank 3 with score 0.3009 and medium confidence. Comparable relationship sets ended in retained_after_support and share signals such as delivery_delay, email_again, retention_risk, support_ticket. The supporting relationship sets are set-2e608f8185f1.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches delivery_delay, email_again, retention_risk, support_ticket more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the growth lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether retained_after_support happened and feed that reviewed result back into the relationship-set evidence."
    },
    {
      "rank": 4,
      "outcome": "account_registered",
      "action": "Ask them to create an account so saved preferences and restock alerts can be used before discounting.",
      "confidence": "Low",
      "score": "0.2652",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Email, search, page A, then the next best purchase action\nDomain: Ecommerce / D2C\nStakeholder: Growth lead\nRuntime query: A person emails in, searches the website, views page A for waterproof jackets, then looks at the product fit and delivery page. What should we do to get them to buy?\n\nGoverned facts:\n- Goal: purchase\n- Primary path: converted_purchase, score 0.6499, high confidence\n- Supporting signals: email_enquiry, page_a, commercial_intent_page, purchase, website_search\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 600; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 4\n- Outcome: account_registered\n- Action: Ask them to create an account so saved preferences and restock alerts can be used before discounting.\n- Confidence: Low\n- Score: 0.2652\n- Evidence summary: Comparable relationship sets ended in account_registered and share signals such as account_benefit, email_enquiry, registration, size_question.\n- Supporting relationship sets: set-a3d51f043f68\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Ask them to create an account so saved preferences and restock alerts can be used before discounting.\n\nWhy this route exists: Fortress returned account_registered at rank 4 with score 0.2652 and low confidence. Comparable relationship sets ended in account_registered and share signals such as account_benefit, email_enquiry, registration, size_question. The supporting relationship sets are set-a3d51f043f68.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches account_benefit, email_enquiry, registration, size_question more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the growth lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether account_registered happened and feed that reviewed result back into the relationship-set evidence."
    }
  ]
}