[{"data":1,"prerenderedAt":230},["ShallowReactive",2],{"/en/workflows/signal-triage-for-trustworthy-decisions":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":211,"tags":212,"integrations":215,"difficulty":217,"author":218,"verified":33,"featured":33,"date":219,"modified":219,"icon":7,"imageSrc":7,"path":220,"alternates":221,"seo":222},"a77b1060-2ca4-4c74-865c-9ae4f772a432","signal-triage-for-trustworthy-decisions","en",null,true,"Signal Triage for Trustworthy Decisions","A guided inbox workflow that turns messy branch metrics, conversations, and attribution into decision-ready checkpoints—before a polished dashboard leads you into a confident mistake.","## How it works\nThis workflow acts like a practical “decision pre-flight check” for branch-level numbers, conversation signals, and attribution claims. It first tries to answer using your Knowledge Base, then guides the requester through a short menu of decision-shaped paths (audit a metric, detect dirty signal, compare branches, etc.).\n\nInstead of chasing perfect cleanliness, it focuses on the failure modes that look fine right up until a meeting goes off the rails: shifting definitions, selective visibility, attribution mirages, and automation used as a substitute for judgment.\n\n## Key features\n- Knowledge Base-first responses to handle common questions quickly, with routing available when the KB can’t help.\n- A decision-focused menu that funnels messy asks into five practical “what should I trust?” paths.\n- Clear, operator-friendly checklists that highlight where polished noise usually hides.\n- Structured escalation to a Decision Support team after the user gets a first-pass reality check.\n\n## Step-by-step\n1. **Trigger:** A new inbound conversation starts the workflow.\n2. **Knowledge Base attempt:** The assistant answers from your KB when possible (definitions, policies, metric owners, FAQ).\n3. **Orientation message:** The assistant sets expectations: this is a quick trust-check, not an academic lecture.\n4. **User chooses a path:** A menu lets the user pick what they’re trying to decide (audit a number, spot dirty signal, automation vs judgment, compare branches, or build signal culture).\n5. **Checklist response:** The workflow sends a tailored set of checks for the selected path.\n6. **Handoff:** The conversation routes to **Decision Support** for follow-up with the user’s context and the checklist they received.\n\n## Setup requirements\n- **Calypso Inbox** enabled for inbound conversations.\n- A populated **Knowledge Base** (recommended: metric definitions, data owners, branch event taxonomy, attribution rules, and known caveats).\n- A configured **Decision Support** department in Calypso routing (no external credentials required).",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":168,"routingEnabled":8,"active":33},"wf_signal_triage_trustworthy_decisions_v1","calypso-public-library","1.0.0",1,[18,34,40,52,60,88,97,104,110,116,121,127,133,139,144,150,161],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":32,"deletable":33,"connectable":33},"node_flow_configs","Workflow settings","flow-configs",[23,24],120,80,{"name":9,"description":26,"tags":27,"triggerType":31},"KB-first decision triage for branch metrics, conversations, and attribution. Routes to Decision Support after delivering practical trust-checks.",[28,29,30],"signal-quality","branch-metrics","decision-making","input","policy",false,{"id":35,"name":36,"type":31,"typeVersion":16,"position":37,"parameters":39,"category":31,"deletable":33,"connectable":8},"node_input","Inbound message",[23,38],220,{},{"id":41,"name":42,"type":43,"typeVersion":16,"position":44,"parameters":46,"category":51,"deletable":8,"connectable":8},"node_kb_policy","Knowledge Base: Signal & metric definitions","knowledge-base-policy",[45,38],360,{"enabled":8,"fallbackToRouting":8,"sticky":33,"stickyMode":47,"activationOpener":48,"personalization":50},"default",{"enabled":8,"instruction":49},"Use the Knowledge Base to answer signal/metric definition questions first. If the user is asking for judgment or a sanity-check, keep it practical: highlight common failure modes and ask one clarifying question when needed. Avoid academic language.",{"useContactName":8},"response",{"id":53,"name":54,"type":55,"typeVersion":16,"position":56,"parameters":58,"category":51,"deletable":8,"connectable":8},"node_intro_text","Set expectations","text-message",[57,38],600,{"text":59},"If you’re about to make a decision off branch numbers, conversation trends, or attribution… good. Let’s do a quick trust-check before the dashboard smiles and the meeting gets loud.\n\nPick what you’re trying to decide and I’ll give you the shortest set of checks that catch the most expensive mistakes.",{"id":61,"name":62,"type":63,"typeVersion":16,"position":64,"parameters":66,"category":51,"deletable":8,"connectable":8},"node_menu","Choose a trust-check path","interactive-message",[65,38],860,{"messageType":67,"headerText":68,"bodyText":69,"footerText":70,"sectionTitle":71,"buttons":72,"ctaDisplayText":71,"ctaUrl":71},"button","Signal triage menu","What do you want to pressure-test?","Tip: avoid confidently wrong calls.","",[73,76,79,82,85],{"id":74,"title":75},"btn_audit_metric","Trust a number",{"id":77,"title":78},"btn_dirty_signal","Spot dirty signal",{"id":80,"title":81},"btn_auto_vs_judgment","Auto vs judgment",{"id":83,"title":84},"btn_compare_branches","Compare branches",{"id":86,"title":87},"btn_signal_culture","Signal culture",{"id":89,"name":90,"type":91,"typeVersion":16,"position":92,"parameters":94,"category":96,"deletable":8,"connectable":8},"node_if_audit_metric","If: Audit metric trust","if",[93,23],1120,{"buttonId":74,"operator":95},"equals","routing",{"id":98,"name":99,"type":55,"typeVersion":16,"position":100,"parameters":102,"category":51,"deletable":8,"connectable":8},"node_audit_metric_text","Checklist: Trustworthy branch numbers",[101,23],1380,{"text":103},"Here’s how to tell “decision-grade” branch numbers from polished noise:\n\n1) Definition lock: did the definition change this month (scope, inclusion rules, time window)?\n2) Denominator sanity: what *could* the branch have done (footfall, leads, staffed hours)? Rates without denominators are confidence cosplay.\n3) Missingness map: which branches fail to report, and do they fail *selectively*?\n4) Incentive check: does anyone get rewarded for this metric? If yes, assume it will be gamed—often unintentionally.\n5) Lag & revision: does the number get revised later? If it does, don’t treat “today” as truth—treat it as a draft.\n\nIf you share the metric name + the decision you’re making, Decision Support can help you confirm whether it’s stable enough to act on.",{"id":105,"name":106,"type":91,"typeVersion":16,"position":107,"parameters":109,"category":96,"deletable":8,"connectable":8},"node_if_dirty_signal","Else-if: Dirty signal",[93,108],240,{"buttonId":77,"operator":95},{"id":111,"name":112,"type":55,"typeVersion":16,"position":113,"parameters":114,"category":51,"deletable":8,"connectable":8},"node_dirty_signal_text","Checklist: Dirty signal warning signs",[101,108],{"text":115},"Dirty signal usually looks clean right up until it’s used. Quick tells:\n\n- Sudden step-change with no operational story (no staffing, hours, promos, outages).\n- One branch “improves” while adjacent branches don’t—especially if processes are shared.\n- The trend matches reporting behavior (new tagging, new script, new form) more than customer reality.\n- A metric improves but its natural “shadow metrics” don’t (e.g., conversion up but revenue, repeat rate, or callbacks don’t budge).\n- The number is unusually smooth. Real life is lumpy.\n\nSend the chart (or the key numbers + dates) and what changed operationally. We’ll route this to Decision Support to sanity-check before it becomes a slide deck fact.",{"id":117,"name":118,"type":91,"typeVersion":16,"position":119,"parameters":120,"category":96,"deletable":8,"connectable":8},"node_if_auto_vs_judgment","Else-if: Automation vs judgment",[93,45],{"buttonId":80,"operator":95},{"id":122,"name":123,"type":55,"typeVersion":16,"position":124,"parameters":125,"category":51,"deletable":8,"connectable":8},"node_auto_vs_judgment_text","Guidance: When to trust automation",[101,45],{"text":126},"Automation is great at consistency. It’s terrible at context. Use this split:\n\nTrust automation when:\n- The input is stable (same definition, same collection method).\n- Errors are cheap and reversible.\n- The decision repeats often (routing, simple thresholds, standard follow-ups).\n\nRequire human judgment when:\n- The metric can be gamed or “managed.”\n- The decision is high-stakes (staffing, closures, major spend).\n- The environment changed (new product, new script, seasonality shift).\n\nRule of thumb: automate the *boring enforcement* of agreed rules; keep humans for exceptions, incentives, and anything that can become a political number.\n\nIf you tell us the decision and what triggers it today, Decision Support can recommend where automation is safe—and where it’s a trap.",{"id":128,"name":129,"type":91,"typeVersion":16,"position":130,"parameters":132,"category":96,"deletable":8,"connectable":8},"node_if_compare_branches","Else-if: Compare branches",[93,131],480,{"buttonId":83,"operator":95},{"id":134,"name":135,"type":55,"typeVersion":16,"position":136,"parameters":137,"category":51,"deletable":8,"connectable":8},"node_compare_branches_text","Checklist: Comparing branches without fooling yourself",[101,131],{"text":138},"Branch comparisons go wrong in repeatable ways. Before ranking anything:\n\n1) Like-for-like: compare on comparable exposure (hours open, staffing, lead volume, local demand).\n2) Mix effects: check whether product/customer mix shifted (a “better” branch may just have easier traffic).\n3) Small numbers: flag branches with low volume—variance will impersonate performance.\n4) Attribution overlap: if multiple channels claim the same outcome, don’t add them up and call it insight.\n5) Process differences: different scripts, tagging, or CRM hygiene can create fake winners.\n\nIf you share the branches, the metric, and the time window, Decision Support can help set a fair comparison frame (and point out the classic traps) before you take action.",{"id":140,"name":141,"type":91,"typeVersion":16,"position":142,"parameters":143,"category":96,"deletable":8,"connectable":8},"node_if_signal_culture","Else-if: Signal culture",[93,57],{"buttonId":86,"operator":95},{"id":145,"name":146,"type":55,"typeVersion":16,"position":147,"parameters":148,"category":51,"deletable":8,"connectable":8},"node_signal_culture_text","Playbook: Build a signal culture (not a slide culture)",[101,57],{"text":149},"Signal culture is simple: make it easier to tell the truth than to defend a metric.\n\nPractical moves that work:\n- One owner per metric definition (and a visible changelog).\n- “Decision note” discipline: every dashboard metric should map to a decision it can trigger.\n- Red-team one metric per month: ask “how could this be wrong?” before asking “how do we raise it?”\n- Reward surfacing uncertainty early. Penalize surprise revisions late.\n- Track data quality as a first-class signal (missingness, revisions, tagging drift).\n\nIf you want, we can route you to Decision Support to help set up a lightweight signal review cadence that produces decisions—not decks.",{"id":151,"name":152,"type":153,"typeVersion":16,"position":154,"parameters":156,"category":160,"deletable":8,"connectable":8},"node_fallback_decision_support","Route to Decision Support","fallback",[155,45],1660,{"handoffMessage":157,"departmentId":158,"departmentName":159},"Got it. I’m routing this to Decision Support for a quick sanity-check. Please share: (1) the decision you’re making, (2) the metric/signal, (3) branches/time window, and (4) anything that changed operationally.","decision-support","Decision Support","terminal",{"id":162,"name":163,"type":153,"typeVersion":16,"position":164,"parameters":166,"category":160,"deletable":8,"connectable":8},"node_fallback_unclear","Route: Clarify request",[101,165],760,{"handoffMessage":167,"departmentId":158,"departmentName":159},"I can help, but I don’t want to guess. I’m routing you to Decision Support—please include the decision you’re making and the signal you’re using so we can pressure-test the right thing.",[169,173,175,177,179,182,184,187,189,191,193,195,197,199,201,203,205,207,209],{"id":170,"source":35,"target":41,"sourceHandle":171,"targetHandle":172,"type":47},"conn_input_to_kb","out","in",{"id":174,"source":41,"target":53,"sourceHandle":171,"targetHandle":172,"type":47},"conn_kb_to_intro",{"id":176,"source":53,"target":61,"sourceHandle":171,"targetHandle":172,"type":47},"conn_intro_to_menu",{"id":178,"source":61,"target":89,"sourceHandle":171,"targetHandle":172,"type":47},"conn_menu_to_if_audit",{"id":180,"source":89,"target":98,"sourceHandle":181,"targetHandle":172,"type":47},"conn_if_audit_true_to_text","true",{"id":183,"source":98,"target":151,"sourceHandle":171,"targetHandle":172,"type":47},"conn_audit_text_to_fallback",{"id":185,"source":89,"target":105,"sourceHandle":186,"targetHandle":172,"type":47},"conn_if_audit_false_to_if_dirty","false",{"id":188,"source":105,"target":111,"sourceHandle":181,"targetHandle":172,"type":47},"conn_if_dirty_true_to_text",{"id":190,"source":111,"target":151,"sourceHandle":171,"targetHandle":172,"type":47},"conn_dirty_text_to_fallback",{"id":192,"source":105,"target":117,"sourceHandle":186,"targetHandle":172,"type":47},"conn_if_dirty_false_to_if_auto",{"id":194,"source":117,"target":122,"sourceHandle":181,"targetHandle":172,"type":47},"conn_if_auto_true_to_text",{"id":196,"source":122,"target":151,"sourceHandle":171,"targetHandle":172,"type":47},"conn_auto_text_to_fallback",{"id":198,"source":117,"target":128,"sourceHandle":186,"targetHandle":172,"type":47},"conn_if_auto_false_to_if_compare",{"id":200,"source":128,"target":134,"sourceHandle":181,"targetHandle":172,"type":47},"conn_if_compare_true_to_text",{"id":202,"source":134,"target":151,"sourceHandle":171,"targetHandle":172,"type":47},"conn_compare_text_to_fallback",{"id":204,"source":128,"target":140,"sourceHandle":186,"targetHandle":172,"type":47},"conn_if_compare_false_to_if_culture",{"id":206,"source":140,"target":145,"sourceHandle":181,"targetHandle":172,"type":47},"conn_if_culture_true_to_text",{"id":208,"source":145,"target":151,"sourceHandle":171,"targetHandle":172,"type":47},"conn_culture_text_to_fallback",{"id":210,"source":140,"target":162,"sourceHandle":186,"targetHandle":172,"type":47},"conn_if_culture_false_to_unclear","automation",[28,29,30,213,214],"data-hygiene","triage",[216],"Calypso Inbox","intermediate","Calypso","2026-07-03T11:03:18.098Z","/en/workflows/signal-triage-for-trustworthy-decisions",{"en":220},{"title":9,"description":223,"ogDescription":224,"twitterDescription":225,"canonicalPath":220,"robots":226,"schemaType":227,"alternates":228},"Guide teams to trust the right branch signals with KB first answers, decision shaped checklists, and a clean handoff to Decision Support.","Stop polished noise from driving confident wrong decisions. This workflow triages branch signals with KB first answers, practical checklists, and Decision Support routing.","Turn messy branch signals into decision ready checkpoints: KB first help, five practical triage paths, and a clean handoff to Decision Support.","index,follow","HowTo",[229],{"hreflang":6,"href":220},1785947667607]