[{"data":1,"prerenderedAt":211},["ShallowReactive",2],{"/en/workflows/branch-evidence-to-decisions-signal-triage-advisor":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":190,"tags":191,"integrations":194,"difficulty":197,"author":198,"verified":33,"featured":33,"date":199,"modified":199,"icon":7,"imageSrc":7,"path":200,"alternates":201,"seo":202},"61642115-0630-48a9-a2c8-50dc21b57575","branch-evidence-to-decisions-signal-triage-advisor","en",null,true,"Branch Evidence to Decisions: Signal Triage Advisor","A decision-focused conversation flow that helps leaders and analysts separate trustworthy branch signals from polished noise, catch dirty data before meetings, and decide when automation is safe to trust.","## How it works\nThis workflow turns “messy signals” into decision-shaped guidance—fast. It starts by searching your Knowledge Base for any internal definitions (metric owners, branch rules, attribution notes), then routes the user into a focused triage menu.\n\nInstead of giving generic data advice, it pushes people toward the questions that prevent confident wrong decisions: which numbers deserve trust, how to spot dirty signal, when automation is safe, what branch comparisons usually get wrong, and how to build a signal culture that ends in decisions—not slides.\n\n## Key features\n- Knowledge Base-first behavior to anchor answers in your internal metric definitions and operating rules\n- A decision-shaped menu that routes users to the exact “guardrail” they need (trust, hygiene, automation, comparisons, culture)\n- Practical, meeting-proof checks to catch polished noise before it becomes a confident narrative\n- Clear next actions (what to verify, what to measure, what to stop pretending is comparable)\n- Optional human handoff path when the user’s situation doesn’t fit the menu\n\n## Step-by-step\n1. **Trigger:** A user starts the conversation (Input).\n2. **Grounding:** The workflow checks the **Knowledge Base Policy** node first to use any internal definitions or standards before routing.\n3. **Triage menu:** The user chooses what they’re deciding via an **Interactive Message** menu.\n4. **Route by intent:** A chain of **If** nodes matches the selected button id.\n5. **Deliver the playbook:** The workflow sends a tailored **Text Message** with practical checks and decision guidance.\n6. **Escalate if needed:** If no option matches, the workflow uses **Fallback** to hand off to a human team.\n\n## Setup requirements\n- **Calypso Knowledge Base** connected and populated with your metric definitions, branch rules, attribution notes, and ownership (recommended, but the workflow still runs without it).\n- No additional credentials are required beyond standard Calypso access.\n- Optional: A configured department for handoff (used by the Fallback node).",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":159,"routingEnabled":8,"active":33},"wf_branch_signal_triage_decision_ready_metrics","calypso-public-library","1.0.0",1,[18,34,40,52,80,89,98,104,110,116,122,128,135,141,148],{"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],80,60,{"name":9,"description":26,"tags":27,"triggerType":31},"Decision-shaped triage for branch signals: trust checks, dirty-signal spotting, automation vs judgment, comparisons/attribution, and signal culture.",[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","Start",[23,38],200,{},{"id":41,"name":42,"type":43,"typeVersion":16,"position":44,"parameters":46,"category":51,"deletable":8,"connectable":8},"node_kb_policy","Knowledge base grounding","knowledge-base-policy",[45,38],320,{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":47,"activationOpener":48,"personalization":50},"ai_sticky_release",{"enabled":8,"instruction":49},"Use our internal metric definitions and branch rules when available. If definitions conflict or are missing, say so plainly and proceed with cautious, decision-first guidance.",{"useContactName":8},"response",{"id":53,"name":54,"type":55,"typeVersion":16,"position":56,"parameters":58,"category":51,"deletable":8,"connectable":8},"node_menu","Choose what you're deciding","interactive-message",[57,38],580,{"messageType":59,"headerText":60,"bodyText":61,"footerText":62,"sectionTitle":63,"buttons":64,"ctaDisplayText":63,"ctaUrl":63},"button","Signal triage","Pick the decision you’re trying to make. I’ll give you the fastest checks to avoid a confident wrong turn.","Tip: Compare branches first to avoid “fairness” traps.","",[65,68,71,74,77],{"id":66,"title":67},"trust_branch_numbers","Trust this metric",{"id":69,"title":70},"spot_dirty_signal","Spot dirty signal",{"id":72,"title":73},"automation_vs_judgment","Auto vs judgment",{"id":75,"title":76},"compare_branches_attr","Branch compare",{"id":78,"title":79},"build_signal_culture","Signal culture",{"id":81,"name":82,"type":83,"typeVersion":16,"position":84,"parameters":86,"category":88,"deletable":8,"connectable":8},"node_if_trust","If: Trust branch numbers","if",[85,38],840,{"buttonId":66,"operator":87},"equals","routing",{"id":90,"name":91,"type":92,"typeVersion":16,"position":93,"parameters":96,"category":51,"deletable":8,"connectable":8},"node_text_trust","Playbook: Which numbers deserve trust","text-message",[94,95],1100,120,{"text":97},"Here’s how to tell “decision-grade” branch numbers from polished noise:\n\n1) **Ownership beats elegance.** Trust metrics with a named owner, clear definition, and a change log. Be wary of beautiful dashboards nobody “owns.”\n\n2) **Denominator sanity check.** Before you argue about a rate, ask: *Who is counted, who is excluded, and did that change last month?* Most branch miracles are denominator drift.\n\n3) **Reconciliation test.** A trustworthy metric can be triangulated (e.g., sales count ↔ deposits ↔ inventory/fulfillment). If it can’t reconcile with anything, treat it like a rumor with formatting.\n\n4) **Latency and revisions.** If the number is revised after close, it’s fine—*as long as that’s documented.* Silent backfills are where confidence goes to die.\n\nIf you tell me the metric name and what decision it will drive (hire, budget, staffing, campaign, hours), I can suggest the minimum checks to run before the meeting.",{"id":99,"name":100,"type":83,"typeVersion":16,"position":101,"parameters":103,"category":88,"deletable":8,"connectable":8},"node_if_dirty","If: Spot dirty signal",[85,102],340,{"buttonId":69,"operator":87},{"id":105,"name":106,"type":92,"typeVersion":16,"position":107,"parameters":108,"category":51,"deletable":8,"connectable":8},"node_text_dirty","Playbook: Dirty signal before it derails decisions",[94,45],{"text":109},"Dirty signal usually looks *plausible*—right until someone bets a quarter on it. Quick red flags:\n\n- **Too smooth to be true.** Perfectly steady trends in messy operations often mean aggregation artifacts or late updates.\n- **Step-changes that align with process changes.** New POS flow, new tagging, new script, new campaign naming—expect breaks in continuity.\n- **One field suddenly ‘improves.’** If conversion spikes but footfall/traffic doesn’t, you may be measuring a new definition of “visit.”\n- **Missingness that’s selective.** If a branch has fewer records only on weekends or only for one rep, it’s not random—it’s behavior or system friction.\n\nMeeting-proof move: ask for **a one-week sample of raw records** (not a chart) and scan for duplicates, gaps, and weird defaults. Charts hide sins; rows confess them.",{"id":111,"name":112,"type":83,"typeVersion":16,"position":113,"parameters":115,"category":88,"deletable":8,"connectable":8},"node_if_auto","If: Automation vs judgment",[85,114],480,{"buttonId":72,"operator":87},{"id":117,"name":118,"type":92,"typeVersion":16,"position":119,"parameters":120,"category":51,"deletable":8,"connectable":8},"node_text_auto","Playbook: When to trust automation",[94,114],{"text":121},"Automation is great at **repeatable** decisions with **bounded risk**. Humans are still required when meaning is changing.\n\nTrust automation when:\n- The decision is frequent (daily/weekly) and reversible.\n- Inputs are stable and defined (same source, same rules).\n- You can measure error cheaply (alerts, spot checks, reconciliation).\n\nRequire human judgment when:\n- The cost of a wrong decision is asymmetric (one bad call hurts more than ten good calls help).\n- The definition is contested (what counts as a lead, a visit, an assisted sale).\n- Branch context matters (staffing outages, local events, system downtime).\n\nPractical rule: **automate the recommendation, not the accountability.** Keep a lightweight override reason (“promo anomaly”, “system outage”, “staffing gap”) so the model doesn’t ‘learn’ your silence.",{"id":123,"name":124,"type":83,"typeVersion":16,"position":125,"parameters":127,"category":88,"deletable":8,"connectable":8},"node_if_compare","If: Compare branches & attribution",[85,126],620,{"buttonId":75,"operator":87},{"id":129,"name":130,"type":92,"typeVersion":16,"position":131,"parameters":133,"category":51,"deletable":8,"connectable":8},"node_text_compare","Playbook: What teams misread in comparisons",[94,132],640,{"text":134},"Branch comparisons fail for boring reasons—then get dressed up as strategy.\n\nCommon misreads:\n1) **Unequal opportunity.** Different footfall, hours, catchment, inventory, or staffing means you’re ranking constraints, not performance.\n2) **Channel leakage.** One branch gets more online-assisted pickups; another gets walk-ins. Attribution moves, not outcomes.\n3) **Different measurement friction.** If one branch follows the script and another doesn’t, you’re measuring compliance—not customer behavior.\n4) **Mix shift disguised as improvement.** A branch ‘improves’ because it served different products/customers, not because it got better.\n\nMake comparisons fair by standardizing one layer at a time:\n- Compare **within** similar cohorts (size, hours, segment).\n- Use **per-opportunity** metrics (per visit, per staffed hour), not just per day.\n- Ask: *If I swapped managers, would the metric still differ?* If yes, it’s structural; don’t moralize it.",{"id":136,"name":137,"type":83,"typeVersion":16,"position":138,"parameters":140,"category":88,"deletable":8,"connectable":8},"node_if_culture","If: Build signal culture",[85,139],760,{"buttonId":78,"operator":87},{"id":142,"name":143,"type":92,"typeVersion":16,"position":144,"parameters":146,"category":51,"deletable":8,"connectable":8},"node_text_culture","Playbook: A signal culture that makes decisions",[94,145],800,{"text":147},"A strong signal culture doesn’t produce more dashboards—it produces fewer surprise meetings.\n\nBuild it with a few habits:\n- **Decision-first framing.** Every metric review starts with: “What decision could change this week because of this?” No decision, no drama.\n- **Definitions are assets.** Maintain a living definition for each key metric: owner, source, exclusions, latency, and known failure modes.\n- **Pre-mortems for metrics.** Ask: “If this metric lies to us, how will it lie?” Write the top 3 ways and add cheap checks.\n- **One gold sample.** Keep a small, audited set of raw examples everyone trusts. When the dashboard argues, the sample is the referee.\n\nWit (and truth): If your culture rewards the *best story*, you’ll get the *best story*. Reward the best **disconfirming evidence** and your decisions will quietly get expensive in a good way.",{"id":149,"name":150,"type":151,"typeVersion":16,"position":152,"parameters":154,"category":158,"deletable":8,"connectable":8},"node_fallback","Handoff to human help","fallback",[94,153],940,{"handoffMessage":155,"departmentId":156,"departmentName":157},"I can hand this to a human for a deeper look (definitions, branch context, and what might be skewing the numbers). Share the metric name, time window, and the decision you’re about to make.","ops-analytics","Operations Analytics","terminal",[160,164,166,168,171,174,176,178,180,182,184,186,188],{"id":161,"source":35,"target":41,"sourceHandle":162,"targetHandle":162,"type":163},"conn_input_to_kb","main","default",{"id":165,"source":41,"target":53,"sourceHandle":162,"targetHandle":162,"type":163},"conn_kb_to_menu",{"id":167,"source":53,"target":81,"sourceHandle":162,"targetHandle":162,"type":163},"conn_menu_to_if_trust",{"id":169,"source":81,"target":90,"sourceHandle":170,"targetHandle":162,"type":163},"conn_if_trust_true_to_text","true",{"id":172,"source":81,"target":99,"sourceHandle":173,"targetHandle":162,"type":163},"conn_if_trust_false_to_if_dirty","false",{"id":175,"source":99,"target":105,"sourceHandle":170,"targetHandle":162,"type":163},"conn_if_dirty_true_to_text",{"id":177,"source":99,"target":111,"sourceHandle":173,"targetHandle":162,"type":163},"conn_if_dirty_false_to_if_auto",{"id":179,"source":111,"target":117,"sourceHandle":170,"targetHandle":162,"type":163},"conn_if_auto_true_to_text",{"id":181,"source":111,"target":123,"sourceHandle":173,"targetHandle":162,"type":163},"conn_if_auto_false_to_if_compare",{"id":183,"source":123,"target":129,"sourceHandle":170,"targetHandle":162,"type":163},"conn_if_compare_true_to_text",{"id":185,"source":123,"target":136,"sourceHandle":173,"targetHandle":162,"type":163},"conn_if_compare_false_to_if_culture",{"id":187,"source":136,"target":142,"sourceHandle":170,"targetHandle":162,"type":163},"conn_if_culture_true_to_text",{"id":189,"source":136,"target":149,"sourceHandle":173,"targetHandle":162,"type":163},"conn_if_culture_false_to_fallback","automation",[28,29,30,192,193],"data-hygiene","attribution",[195,196],"Calypso Knowledge Base","Calypso Inbox","intermediate","Calypso","2026-07-28T11:03:58.735Z","/en/workflows/branch-evidence-to-decisions-signal-triage-advisor",{"en":200},{"title":203,"description":204,"ogDescription":205,"twitterDescription":206,"canonicalPath":200,"robots":207,"schemaType":208,"alternates":209},"Branch Signal Triage for Decision Ready Metrics","Guide teams to trust the right branch numbers, spot dirty signals early, and decide when automation is safe—via a routed chat advisor.","A routed advisor that helps leaders separate trustworthy branch metrics from polished noise, catch dirty signals before meetings, and know when to trust automation.","Turn messy branch signals into decision ready guidance: what to trust, what’s dirty, what comparisons lie, and when automation needs human judgment.","index,follow","HowTo",[210],{"hreflang":6,"href":200},1785947666149]