[{"data":1,"prerenderedAt":211},["ShallowReactive",2],{"/en/workflows/signal-triage-coach-for-branch-decisions":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":192,"tags":193,"integrations":196,"difficulty":199,"author":197,"verified":33,"featured":33,"date":200,"modified":200,"icon":7,"imageSrc":7,"path":201,"alternates":202,"seo":203},"9342832e-af0b-4626-b384-4cc336323c72","signal-triage-coach-for-branch-decisions","en",null,true,"Signal Triage Coach for Branch Decisions","An interactive decision-support flow that helps teams quickly judge which branch signals to trust, spot dirty data early, and choose when automation is safe vs when human judgment must intervene.","## How it works\nThis workflow is a lightweight “decision coach” you can drop into a chat channel to keep branch reviews from getting seduced by polished noise. It guides a manager or analyst through quick, decision-shaped checks: whether a number deserves trust, how to spot dirty signal before the meeting, and when automation is a gift vs a liability.\n\nIt’s designed for the moment right before confidence hardens. Users pick the situation they’re in, get a practical checklist (with a little edge), and can loop back to run another check or request a handoff to a human.\n\n## Key features\n- Knowledge-first behavior: the flow can answer using your existing knowledge base before routing users to guided options.\n- Button-based menu for fast triage across common decision risks (branch metrics, dirty signals, attribution, automation vs judgment).\n- Practical “trust tests” that focus on what fails first (definitions, timing, incentives, missingness), not academic theory.\n- Loop-back design so users can run multiple checks in one session without restarting.\n- Built-in human handoff path when the situation is messy or high-stakes.\n\n## Step-by-step\n1. **Trigger:** A user starts the workflow (e.g., from web chat).\n2. **Knowledge base check:** The workflow attempts to handle the user’s intent with your knowledge base policy.\n3. **Menu:** The user chooses what they need:\n   - **Which branch numbers deserve trust**\n   - **Spot dirty signal before the meeting**\n   - **Automation vs human judgment**\n   - **Comparing branches & attribution**\n   - **Build a signal culture (less slide theater)**\n   - **Talk to a human**\n4. **Guidance:** The workflow sends a targeted, plain-English checklist for the selected scenario.\n5. **Loop or handoff:** After guidance, the workflow returns to the menu so the user can run another check—or routes to **Human handoff** if requested.\n\n## Setup requirements\n- No credentials required.\n- (Optional) Configure your Calypso knowledge base content so the **Knowledge base check** can answer common signal/metric questions before the menu routing.\n- If you use the **Talk to a human** option, ensure your Calypso routing/handoff settings and department naming match your operational process.",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":161,"routingEnabled":8,"active":33},"wf_signal_triage_coach_v1","calypso-public-library","1.0.0",1,[18,34,41,53,83,92,98,103,109,115,121,129,135,141,147,153],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":32,"deletable":33,"connectable":33},"cfg","Flow settings","flow-configs",[23,24],-520,-40,{"name":9,"description":26,"tags":27,"triggerType":31},"Interactive decision-support coach to validate branch signals, spot dirty data, and choose automation vs judgment.",[28,29,30],"signal-triage","branch-metrics","decision-support","input","policy",false,{"id":35,"name":36,"type":31,"typeVersion":16,"position":37,"parameters":40,"category":31,"deletable":33,"connectable":8},"in","Start",[38,39],-240,120,{},{"id":42,"name":43,"type":44,"typeVersion":16,"position":45,"parameters":47,"category":52,"deletable":8,"connectable":8},"kb","Knowledge base check","knowledge-base-policy",[46,39],-20,{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":48,"activationOpener":49,"personalization":51},"default",{"enabled":33,"instruction":50},"",{"useContactName":8},"response",{"id":54,"name":55,"type":56,"typeVersion":16,"position":57,"parameters":59,"category":52,"deletable":8,"connectable":8},"menu","Choose a signal check","interactive-message",[58,39],220,{"messageType":60,"headerText":61,"bodyText":62,"footerText":63,"sectionTitle":50,"buttons":64,"ctaDisplayText":50,"ctaUrl":50},"button","Signal triage menu","Pick what you’re deciding. I’ll help you avoid the classic: a clean-looking number that’s quietly wrong.","High-stakes decision? Use Human handoff early.",[65,68,71,74,77,80],{"id":66,"title":67},"branch_trust","Trust branch numbers",{"id":69,"title":70},"dirty_signal","Spot dirty signal",{"id":72,"title":73},"auto_vs_human","Auto vs human",{"id":75,"title":76},"compare_attrib","Compare & attribute",{"id":78,"title":79},"signal_culture","Signal culture",{"id":81,"title":82},"human_handoff","Human handoff",{"id":84,"name":85,"type":86,"typeVersion":16,"position":87,"parameters":89,"category":91,"deletable":8,"connectable":8},"if_branch_trust","If: branch trust","if",[88,24],460,{"buttonId":66,"operator":90},"equals","routing",{"id":93,"name":94,"type":86,"typeVersion":16,"position":95,"parameters":97,"category":91,"deletable":8,"connectable":8},"if_dirty_signal","If: dirty signal",[88,96],40,{"buttonId":69,"operator":90},{"id":99,"name":100,"type":86,"typeVersion":16,"position":101,"parameters":102,"category":91,"deletable":8,"connectable":8},"if_auto_vs_human","If: automation vs judgment",[88,39],{"buttonId":72,"operator":90},{"id":104,"name":105,"type":86,"typeVersion":16,"position":106,"parameters":108,"category":91,"deletable":8,"connectable":8},"if_compare_attrib","If: compare branches & attribution",[88,107],200,{"buttonId":75,"operator":90},{"id":110,"name":111,"type":86,"typeVersion":16,"position":112,"parameters":114,"category":91,"deletable":8,"connectable":8},"if_signal_culture","If: signal culture",[88,113],280,{"buttonId":78,"operator":90},{"id":116,"name":117,"type":86,"typeVersion":16,"position":118,"parameters":120,"category":91,"deletable":8,"connectable":8},"if_handoff","If: human handoff",[88,119],360,{"buttonId":81,"operator":90},{"id":122,"name":123,"type":124,"typeVersion":16,"position":125,"parameters":127,"category":52,"deletable":8,"connectable":8},"msg_branch_trust","Trust test for branch numbers","text-message",[126,24],740,{"text":128},"Here’s the fastest way to tell “decision-grade” branch numbers from polished noise:\n\n1) Definition lock: Can two people compute it the same way? If not, it’s a story, not a metric.\n2) Denominator sanity: What changed underneath (traffic, staffing, hours, customer mix)? Most ‘improvements’ ride the denominator.\n3) Timing: Is it leading, lagging, or just late? A lagging metric is fine—if you don’t pretend it predicts next week.\n4) Incentives: If someone gets rewarded for it, assume it will be gamed (often unintentionally).\n5) Missingness: Ask “who’s not counted?” The absent cases are usually the problem cases.\n\nIf you can’t answer #1–#3 in one minute, don’t argue the fourth decimal place. Ask for the definition and the data cut first.",{"id":130,"name":131,"type":124,"typeVersion":16,"position":132,"parameters":133,"category":52,"deletable":8,"connectable":8},"msg_dirty_signal","Spot dirty signal before the meeting",[126,96],{"text":134},"Dirty signal usually looks perfectly reasonable—until you rely on it. Quick pre-meeting checks:\n\n• Sudden step-change? Check for a tracking/definition change before you celebrate.\n• Too smooth? Real operations are lumpy. Perfect lines often mean backfilled data or averaging that hides failure.\n• “Everyone improved” at once? That’s often policy/reporting drift, not performance.\n• Outliers: Are they real events (storms, staffing gaps) or logging glitches? Treat them as clues, not trash.\n• Reconciliation: Do totals match across reports? If two dashboards disagree, the faster one isn’t “more real.”\n\nRule of thumb: if a metric can’t survive three dumb questions, it shouldn’t survive a confident meeting.",{"id":136,"name":137,"type":124,"typeVersion":16,"position":138,"parameters":139,"category":52,"deletable":8,"connectable":8},"msg_auto_vs_human","When to trust automation vs judgment",[126,39],{"text":140},"Automation is great at consistency; it’s terrible at context.\n\nTrust automation when:\n• The decision is frequent, reversible, and bounded (clear thresholds, clear outcomes).\n• Inputs are stable and well-instrumented (same definitions across branches).\n• You’ve measured error cost (false positives/negatives) and it’s acceptable.\n\nRequire human judgment when:\n• The decision is high-stakes or one-way (reputation, safety, compliance, closures).\n• The environment changed (new campaign, staffing model, seasonality shift).\n• The metric is proxy-heavy (attribution, ‘intent’, ‘quality’) or easy to game.\n\nBest practice: automate the recommendation and force a human to confirm the exception cases. That’s where reality lives.",{"id":142,"name":143,"type":124,"typeVersion":16,"position":144,"parameters":145,"category":52,"deletable":8,"connectable":8},"msg_compare_attrib","Comparing branches & attribution (without self-deception)",[126,107],{"text":146},"Branch comparisons fail in the same three places:\n\n1) Mix effects: Branch A isn’t Branch B. Customer mix, hours, staffing, and local demand will clown your leaderboard.\n2) Exposure: If one branch had more opportunities (traffic, leads, calls), raw outcomes lie. Normalize thoughtfully.\n3) Attribution mirages: If a conversation or campaign “gets credit,” check whether it simply touches the easiest cases.\n\nPractical comparison moves:\n• Compare like with like (segments, time windows, similar staffing).\n• Use paired metrics: outcome + volume + a guardrail (e.g., conversion + complaints).\n• Look for reversals: if the rank flips when you change one assumption, the metric isn’t stable enough to bet on.\n\nIf the conclusion only works in one chart, it’s not a conclusion yet.",{"id":148,"name":149,"type":124,"typeVersion":16,"position":150,"parameters":151,"category":52,"deletable":8,"connectable":8},"msg_signal_culture","Build a signal culture that makes decisions",[126,113],{"text":152},"A good signal culture isn’t “more metrics.” It’s fewer arguments and faster decisions.\n\nDo this:\n• Define “decision-grade” up front: owner, definition, refresh cadence, known blind spots.\n• Keep a small metric menu: 3–5 per decision, with one guardrail metric that prevents gaming.\n• Track changes to definitions like you track code changes. Quiet tweaks create loud confusion.\n• Celebrate someone who kills a bad metric. That’s quality control, not negativity.\n\nAvoid this:\n• Dashboard sprawl (“we’ll know it when we see it”). That’s how slide decks win and truth loses.\n\nIf your weekly review can’t name what changed, why it changed, and what you’ll do next—your signals are decor.",{"id":154,"name":82,"type":155,"typeVersion":16,"position":156,"parameters":157,"category":160,"deletable":8,"connectable":8},"handoff","fallback",[126,119],{"handoffMessage":158,"departmentId":50,"departmentName":159},"Got it. This sounds like a case where context matters. I’m handing this to a teammate for a human review.","Operations","terminal",[162,165,167,169,171,173,175,177,179,182,184,186,188,190],{"id":163,"source":35,"target":42,"sourceHandle":164,"targetHandle":35,"type":48},"c1","out",{"id":166,"source":42,"target":54,"sourceHandle":164,"targetHandle":35,"type":48},"c2",{"id":168,"source":54,"target":84,"sourceHandle":164,"targetHandle":35,"type":48},"c3",{"id":170,"source":54,"target":93,"sourceHandle":164,"targetHandle":35,"type":48},"c4",{"id":172,"source":54,"target":99,"sourceHandle":164,"targetHandle":35,"type":48},"c5",{"id":174,"source":54,"target":104,"sourceHandle":164,"targetHandle":35,"type":48},"c6",{"id":176,"source":54,"target":110,"sourceHandle":164,"targetHandle":35,"type":48},"c7",{"id":178,"source":54,"target":116,"sourceHandle":164,"targetHandle":35,"type":48},"c8",{"id":180,"source":84,"target":122,"sourceHandle":181,"targetHandle":35,"type":48},"c9","true",{"id":183,"source":93,"target":130,"sourceHandle":181,"targetHandle":35,"type":48},"c10",{"id":185,"source":99,"target":136,"sourceHandle":181,"targetHandle":35,"type":48},"c11",{"id":187,"source":104,"target":142,"sourceHandle":181,"targetHandle":35,"type":48},"c12",{"id":189,"source":110,"target":148,"sourceHandle":181,"targetHandle":35,"type":48},"c13",{"id":191,"source":116,"target":154,"sourceHandle":181,"targetHandle":35,"type":48},"c14","automation",[28,29,30,194,195],"data-hygiene","automation-judgment",[197,198],"Calypso","Web chat","intermediate","2026-05-24T11:03:46.468Z","/en/workflows/signal-triage-coach-for-branch-decisions",{"en":201},{"title":9,"description":204,"ogDescription":205,"twitterDescription":206,"canonicalPath":201,"robots":207,"schemaType":208,"alternates":209},"Guide teams to trust the right branch signals, catch dirty data early, and choose when automation is safe vs when humans should decide.","An interactive chat workflow that stress tests branch metrics, spots dirty signals, and prevents confident wrong decisions—before the meeting does damage.","A practical signal triage flow: trust tests for branch numbers, dirty signal spotting, automation vs judgment, and cleaner comparisons without fake certainty.","index,follow","HowTo",[210],{"hreflang":6,"href":201},1780761213372]