[{"data":1,"prerenderedAt":228},["ShallowReactive",2],{"/en/workflows/branch-signal-triage-trust-noise-and-next-actions":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":210,"tags":211,"integrations":213,"difficulty":215,"author":214,"verified":34,"featured":34,"date":216,"modified":216,"icon":7,"imageSrc":7,"path":217,"alternates":218,"seo":219},"f4ed6c1a-569a-4792-aec2-7dda8a7bb13d","branch-signal-triage-trust-noise-and-next-actions","en",null,true,"Branch Signal Triage: Trust, Noise, and Next Actions","A decision-first coaching flow that helps teams spot polished noise, detect dirty signals, and choose when to trust automation vs human judgment—before the meeting goes off the rails.","## How it works\nThis workflow turns “we have numbers” into “we have decision-ready evidence.” It quickly routes a teammate to the right kind of signal check—branch metrics, messy conversation evidence, attribution comparisons, or the classic automation-vs-judgment call.\n\nIt’s designed for the real failure mode: data that looks clean and persuasive right up until a team makes a confident wrong decision. Each path gives crisp guidance on what to trust, what to measure next, and what usually breaks first.\n\n## Key features\n- One-tap menu to route people to the right signal check (no wandering into a dashboard maze)\n- A knowledge-base-first policy that answers from your existing guidance before routing\n- Practical, decision-shaped checklists focused on branch comparisons, conversations, and attribution\n- Clear escalation path to a human owner when the signal is too risky to “just automate”\n\n## Step-by-step\n1. **Trigger:** A user starts the workflow.\n2. **Knowledge base pass:** The workflow attempts to answer using your knowledge base guidance first.\n3. **Pick a signal check:** The user chooses the situation they’re in (e.g., branch numbers, dirty signal detection, automation vs judgment, attribution comparison, or signal culture).\n4. **Receive targeted guidance:** The workflow sends a focused checklist and next actions for the chosen decision context.\n5. **Escalate when needed:** If the user selects “Bring in a human,” the workflow hands off to the designated department.\n\n## Setup requirements\n- No external credentials are required.\n- (Optional) Configure your Calypso Knowledge Base content so the knowledge-base step can answer with your house definitions (e.g., metric definitions, attribution rules, branch reporting caveats).\n- (Optional) Confirm the handoff department exists in your Calypso routing setup (used for “Bring in a human”).",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":175,"routingEnabled":8,"active":34},"wf_branch_signal_triage_v1","calypso-public-library","1.0.0",1,[18,35,42,54,86,96,104,110,116,122,128,134,140,146,152,158,168],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":33,"deletable":34,"connectable":34},"node_flow_configs","Workflow settings","flow-configs",[23,24],80,60,{"name":9,"description":26,"tags":27,"triggerType":32},"Decision-first triage to detect polished noise, dirty signal, and misread branch/attribution comparisons. Includes optional human handoff.",[28,29,30,31],"signal-hygiene","branch-metrics","decision-systems","attribution","input","policy",false,{"id":36,"name":37,"type":32,"typeVersion":16,"position":38,"parameters":41,"category":32,"deletable":34,"connectable":8},"node_input","Start",[39,40],120,220,{},{"id":43,"name":44,"type":45,"typeVersion":16,"position":46,"parameters":48,"category":53,"deletable":8,"connectable":8},"node_kb_policy","Knowledge base guidance","knowledge-base-policy",[47,40],360,{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":49,"activationOpener":50,"personalization":52},"default",{"enabled":8,"instruction":51},"Use the knowledge base to answer first. If the user’s request is better served by triage, continue to the menu. Keep the tone practical and decision-focused, not academic.",{"useContactName":8},"response",{"id":55,"name":56,"type":57,"typeVersion":16,"position":58,"parameters":60,"category":53,"deletable":8,"connectable":8},"node_menu","Decision menu","interactive-message",[59,40],620,{"messageType":61,"headerText":62,"bodyText":63,"footerText":64,"sectionTitle":65,"buttons":66,"ctaDisplayText":85,"ctaUrl":85},"button","Signal triage","What kind of decision are you trying to make right now? Pick one and I’ll give you the *right* checks (the ones that prevent confident wrong decisions).","Tip: Choose the slide you trust most.","Choose a path",[67,70,73,76,79,82],{"id":68,"title":69},"branch_numbers_trust","Trust branch numbers",{"id":71,"title":72},"spot_dirty_signal","Spot dirty signal",{"id":74,"title":75},"automation_vs_judgment","Auto vs judgment",{"id":77,"title":78},"compare_branches_attribution","Branch + attribution",{"id":80,"title":81},"build_signal_culture","Signal culture",{"id":83,"title":84},"bring_in_human","Human handoff","",{"id":87,"name":88,"type":89,"typeVersion":16,"position":90,"parameters":93,"category":95,"deletable":8,"connectable":8},"node_if_branch_trust","If: Branch numbers trust","if",[91,92],880,140,{"buttonId":68,"operator":94},"equals","routing",{"id":97,"name":98,"type":99,"typeVersion":16,"position":100,"parameters":102,"category":53,"deletable":8,"connectable":8},"node_txt_branch_trust","Branch numbers: what to trust","text-message",[101,92],1140,{"text":103},"Here’s the branch-number reality check (the stuff that saves you from polished noise):\n\n1) **Trust trends over single points.** One great week is often a reporting artifact in a nice outfit.\n2) **Ask “what would move this number without reality changing?”** Backfills, new tagging, staffing changes, cutoff times, and “fixed” definitions.\n3) **Prefer counts tied to a verifiable event.** E.g., appointments attended > appointments booked; funded > approved; resolved > contacted.\n4) **Check denominator drift.** Branch comparisons usually break because the denominator quietly changed (traffic mix, lead quality, hours, catchment).\n5) **Look for a ‘shadow metric’.** If conversion jumps but cancellations, refunds, rework, or time-to-serve doesn’t budge… it’s often measurement, not performance.\n\nIf you tell me the metric and the decision it will drive, I’ll suggest the fastest falsification test.",{"id":105,"name":106,"type":89,"typeVersion":16,"position":107,"parameters":109,"category":95,"deletable":8,"connectable":8},"node_if_dirty_signal","If: Spot dirty signal",[91,108],240,{"buttonId":71,"operator":94},{"id":111,"name":112,"type":99,"typeVersion":16,"position":113,"parameters":114,"category":53,"deletable":8,"connectable":8},"node_txt_dirty_signal","Dirty signal: early warning checks",[101,108],{"text":115},"Dirty signal usually looks *reasonable*—that’s the problem. Quick checks before the meeting:\n\n- **Too smooth:** If a line is perfectly monotonic, ask what got rounded, bucketed, or imputed.\n- **Too convenient:** If the metric moves exactly when the team wanted it to, verify instrumentation and definitions around that date.\n- **Missing bad news:** If there’s no increase in exceptions, complaints, reversals, or delays while output improves, assume you’re not seeing the full system.\n- **Category wobble:** A spike in “Other/Unknown” is a silent data fire.\n- **Selection bias disguised as targeting:** If quality improved after “better filtering,” verify you didn’t just change who gets counted.\n\nRule of thumb: if nobody can explain the *failure modes* of a metric, you’re not holding a metric—you’re holding a prop.",{"id":117,"name":118,"type":89,"typeVersion":16,"position":119,"parameters":121,"category":95,"deletable":8,"connectable":8},"node_if_auto_vs_human","If: Automation vs judgment",[91,120],340,{"buttonId":74,"operator":94},{"id":123,"name":124,"type":99,"typeVersion":16,"position":125,"parameters":126,"category":53,"deletable":8,"connectable":8},"node_txt_auto_vs_human","Automation vs judgment: decision rules",[101,120],{"text":127},"When to trust automation vs when you still need a human:\n\n**Trust automation when:**\n- The input is **observable and consistent** (timestamps, system events, clear states).\n- The cost of a wrong call is **small and reversible**.\n- The rule is stable and you can **audit outcomes**.\n\n**Pull in human judgment when:**\n- The situation is **rare, novel, or high-stakes** (tail risk lives here).\n- The signal is a **proxy** (sentiment, intent, “quality”) and definitions drift.\n- Incentives are strong enough that people can “game” the metric.\n\nA practical compromise: automate the *default*, require humans for **exceptions**, and measure the exception rate—spikes often reveal upstream signal rot.",{"id":129,"name":130,"type":89,"typeVersion":16,"position":131,"parameters":133,"category":95,"deletable":8,"connectable":8},"node_if_compare_attr","If: Compare branches & attribution",[91,132],440,{"buttonId":77,"operator":94},{"id":135,"name":136,"type":99,"typeVersion":16,"position":137,"parameters":138,"category":53,"deletable":8,"connectable":8},"node_txt_compare_attr","Comparisons: what teams misread",[101,132],{"text":139},"Branch + attribution comparisons fail in the same predictable ways:\n\n1) **Different funnels, same label.** “Leads” aren’t comparable if sources, eligibility, or follow-up rules differ.\n2) **Timing is not neutral.** Branch A closes faster; Branch B looks worse only because the window is too short.\n3) **Attribution tells a story, not the truth.** Last-touch is great at rewarding whoever shows up late.\n4) **Mix shift masquerading as performance.** One branch gets harder cases, different hours, or different staffing.\n5) **Small numbers lie loudly.** A few conversions can swing rates; use counts and confidence, not just percentages.\n\nIf you must rank branches: start with *within-branch change*, then explain differences with mix, timing, and process—not vibes.",{"id":141,"name":142,"type":89,"typeVersion":16,"position":143,"parameters":145,"category":95,"deletable":8,"connectable":8},"node_if_signal_culture","If: Build signal culture",[91,144],540,{"buttonId":80,"operator":94},{"id":147,"name":148,"type":99,"typeVersion":16,"position":149,"parameters":150,"category":53,"deletable":8,"connectable":8},"node_txt_signal_culture","Signal culture: make decisions happen",[101,144],{"text":151},"A signal culture that actually helps decisions (not just slides):\n\n- **Make definitions boring and enforced.** Ambiguity is where bad decisions hide.\n- **Promote “show me the instrument.”** A metric without lineage (who logs it, when, how it changes) is a rumor with a chart.\n- **Reward disconfirming evidence.** If people get punished for being the messenger, you’ll only hear good news.\n- **Decide with ‘next best test,’ not ‘perfect dataset.’** The goal is reducing uncertainty, not winning a formatting contest.\n- **Document the failure modes.** For every key metric: how it can be wrong, how you’d notice, and who owns the fix.\n\nIf you want, share the decision you’re trying to unlock and what evidence you have—we’ll shape a minimal, trustworthy signal set.",{"id":153,"name":154,"type":89,"typeVersion":16,"position":155,"parameters":157,"category":95,"deletable":8,"connectable":8},"node_if_handoff","If: Bring in a human",[91,156],640,{"buttonId":83,"operator":94},{"id":159,"name":160,"type":161,"typeVersion":16,"position":162,"parameters":163,"category":167,"deletable":8,"connectable":8},"node_fallback_handoff","Handoff to analytics","fallback",[101,156],{"handoffMessage":164,"departmentId":165,"departmentName":166},"Got it. This looks like a case where human judgment is cheaper than a confident wrong decision. I’m handing this to the Analytics team. Please include: the decision at stake, the metric(s), the time window, and what changed recently.","dept_analytics","Analytics","terminal",{"id":169,"name":170,"type":161,"typeVersion":16,"position":171,"parameters":173,"category":167,"deletable":8,"connectable":8},"node_fallback_default","Default handoff",[101,172],740,{"handoffMessage":174,"departmentId":165,"departmentName":166},"I couldn’t match that selection cleanly. I’m handing this to the Analytics team so you don’t lose momentum. Share the decision, the metric, and the branch/time window.",[176,180,182,184,187,190,192,194,196,198,200,202,204,206,208],{"id":177,"source":36,"target":43,"sourceHandle":178,"targetHandle":179,"type":49},"conn_input_to_kb","out","in",{"id":181,"source":43,"target":55,"sourceHandle":178,"targetHandle":179,"type":49},"conn_kb_to_menu",{"id":183,"source":55,"target":87,"sourceHandle":178,"targetHandle":179,"type":49},"conn_menu_to_if1",{"id":185,"source":87,"target":97,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if1_true_to_txt1","true",{"id":188,"source":87,"target":105,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if1_false_to_if2","false",{"id":191,"source":105,"target":111,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if2_true_to_txt2",{"id":193,"source":105,"target":117,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if2_false_to_if3",{"id":195,"source":117,"target":123,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if3_true_to_txt3",{"id":197,"source":117,"target":129,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if3_false_to_if4",{"id":199,"source":129,"target":135,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if4_true_to_txt4",{"id":201,"source":129,"target":141,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if4_false_to_if5",{"id":203,"source":141,"target":147,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if5_true_to_txt5",{"id":205,"source":141,"target":153,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if5_false_to_if6",{"id":207,"source":153,"target":159,"sourceHandle":186,"targetHandle":179,"type":49},"conn_if6_true_to_handoff",{"id":209,"source":153,"target":169,"sourceHandle":189,"targetHandle":179,"type":49},"conn_if6_false_to_default_handoff","automation",[28,29,30,31,212],"research-triage",[214],"Calypso","intermediate","2026-04-01T11:04:25.137Z","/en/workflows/branch-signal-triage-trust-noise-and-next-actions",{"en":217},{"title":220,"description":221,"ogDescription":222,"twitterDescription":223,"canonicalPath":217,"robots":224,"schemaType":225,"alternates":226},"Branch Signal Triage for Better Decisions","Route teams to the right signal check—trustworthy branch numbers, dirty signal spotting, automation vs judgment, and attribution comparisons.","A practical decision triage: identify trustworthy branch numbers, detect dirty signals early, and know when automation needs human judgment.","Stop confident wrong decisions. Triage branch signals, messy conversation evidence, and attribution comparisons—with a clear human handoff when needed.","index,follow","HowTo",[227],{"hreflang":6,"href":217},1775310170347]