[{"data":1,"prerenderedAt":234},["ShallowReactive",2],{"/en/workflows/branch-signal-decision-radar":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":214,"tags":215,"integrations":219,"difficulty":221,"author":222,"verified":33,"featured":33,"date":223,"modified":223,"icon":7,"imageSrc":7,"path":224,"alternates":225,"seo":226},"52be556a-9c37-43e6-9d06-9160fe64784b","branch-signal-decision-radar","en",null,true,"Branch Signal Decision Radar","Turn messy branch numbers, conversations, and local events into decision-ready guidance—before polished noise turns into confident mistakes.","## How it works\nThis workflow acts like a practical decision “spotter” for branch leaders: it helps you quickly pressure-test branch numbers, conversation signals, and attribution stories before they harden into a confident (wrong) meeting narrative.\n\nIt uses a knowledge-guided coaching step to keep advice consistent, then offers a menu of decision-shaped prompts—so teams can choose the situation they’re in (trusting branch metrics, spotting dirty signal, automation vs. judgment, comparing branches, etc.) and get crisp, usable guidance.\n\n## Key features\n- Knowledge-guided coaching tone to keep responses grounded and non-academic\n- Button-based menu so users self-triage into the right decision scenario fast\n- Separate paths for the most common failure modes: “polished noise,” dirty signal, attribution traps, and branch comparisons\n- Explicit guidance on when automation is safe—and when human judgment must stay in the loop\n- Optional handoff path for cases that need a human review\n\n## Step-by-step\n1. **Trigger:** The workflow starts when a user begins a conversation (Input).\n2. **Knowledge guidance:** A **Knowledge Base Policy** sets the coaching style and keeps answers consistent across paths.\n3. **Choose a scenario:** An **Interactive Message** asks what the user is trying to decide and provides buttons.\n4. **Route by choice:** **IF** nodes route the user to the matching scenario.\n5. **Get decision-ready guidance:** A **Text Message** delivers a concise checklist and “watch-outs” tailored to the selected scenario.\n6. **Escalate if needed:** If the user chooses “Bring in a human,” the workflow sends a **Fallback** handoff message.\n\n## Setup requirements\n- Calypso messaging channel that supports interactive buttons/lists for the menu step\n- A published Knowledge Base connected to Calypso (recommended for best results)\n- No external app credentials are required for this workflow",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":179,"routingEnabled":8,"active":33},"wf_branch_signal_decision_radar_v1","calypso-public-library","1.0.0",1,[18,34,40,51,85,95,103,109,115,121,127,133,139,145,151,157,163,169],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":32,"deletable":33,"connectable":33},"node_flow_configs","Flow Configs","flow-configs",[23,24],-240,40,{"name":9,"description":26,"tags":27,"triggerType":31},"Decision-shaped coaching to help teams trust the right branch signals, spot dirty data, and know when automation needs human judgment.",[28,29,30],"signal-quality","decision-systems","branch-ops","input","policy",false,{"id":35,"name":36,"type":31,"typeVersion":16,"position":37,"parameters":39,"category":31,"deletable":33,"connectable":8},"node_input","Input",[23,38],220,{},{"id":41,"name":42,"type":43,"typeVersion":16,"position":44,"parameters":45,"category":50,"deletable":8,"connectable":8},"node_kb_policy","Decision Coaching Policy","knowledge-base-policy",[24,38],{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":46,"activationOpener":47,"personalization":49},"default",{"enabled":8,"instruction":48},"Coach like an experienced operations advisor. Be practical, slightly witty, and decision-first. Assume bad data often looks fine until a team confidently acts on it. Give checklists, tests, and common failure modes. Avoid academic tone and avoid vague thought leadership.",{"useContactName":33},"response",{"id":52,"name":53,"type":54,"typeVersion":16,"position":55,"parameters":57,"category":50,"deletable":8,"connectable":8},"node_menu","Choose your situation","interactive-message",[56,38],320,{"messageType":58,"headerText":59,"bodyText":60,"footerText":61,"sectionTitle":62,"buttons":63,"ctaDisplayText":62,"ctaUrl":62},"button","Branch signal decision radar","What are you trying to decide right now? Pick the closest match and I’ll give you a tight set of checks (the kind that save you from confident wrong turns).","Tip: If the number looks too clean, that’s usually the plot.","",[64,67,70,73,76,79,82],{"id":65,"title":66},"trust_branch_numbers","Numbers to trust",{"id":68,"title":69},"spot_dirty_signal","Spot dirty signal",{"id":71,"title":72},"automation_vs_judgment","Auto vs judgment",{"id":74,"title":75},"messy_evidence","Messy to insight",{"id":77,"title":78},"compare_branches","Compare branches",{"id":80,"title":81},"signal_culture","Signal culture",{"id":83,"title":84},"bring_human","Human review",{"id":86,"name":87,"type":88,"typeVersion":16,"position":89,"parameters":92,"category":94,"deletable":8,"connectable":8},"if_trust_branch_numbers","IF: Which numbers to trust","if",[90,91],580,60,{"buttonId":65,"operator":93},"equals","routing",{"id":96,"name":97,"type":98,"typeVersion":16,"position":99,"parameters":101,"category":50,"deletable":8,"connectable":8},"msg_trust_branch_numbers","Trusted vs. polished noise","text-message",[100,91],860,{"text":102},"Here’s the fast way to sort *decision-grade* branch numbers from polished noise:\n\n**Numbers you can usually trust (or quickly validate):**\n- **Counted at the moment of truth:** deposits posted, pickups completed, appointments attended.\n- **Hard to “game” locally:** tied to system constraints (inventory movement, teller reconciliation, verified IDs).\n- **Has a clear denominator:** conversion rate with an agreed base (walk-ins, qualified leads, scheduled appointments).\n\n**Numbers that love to lie (especially when targets exist):**\n- **Self-reported outcomes** (\"we contacted 200 customers\") without proof of contact quality.\n- **Vanity activity** (calls placed, emails sent) without disposition integrity.\n- **Composite scores** with changing definitions or hidden weights.\n\n**Three trust tests (use these before the meeting):**\n1) **Definition test:** Did the definition change this month? If yes, treat trendlines as suspect.\n2) **Friction test:** Could a branch hit the number without doing the real work? If yes, it’s a proxy, not an outcome.\n3) **Reconciliation test:** Can you reconcile to an independent source (core banking, appointment system, POS, ticket closure)? If not, it’s a story.\n\nIf you tell me the metric (and where it comes from), I’ll tell you what it’s most likely measuring *by accident*.",{"id":104,"name":105,"type":88,"typeVersion":16,"position":106,"parameters":108,"category":94,"deletable":8,"connectable":8},"if_spot_dirty_signal","IF: Spot dirty signal fast",[90,107],160,{"buttonId":68,"operator":93},{"id":110,"name":111,"type":98,"typeVersion":16,"position":112,"parameters":113,"category":50,"deletable":8,"connectable":8},"msg_spot_dirty_signal","Dirty signal early warnings",[100,107],{"text":114},"Dirty signal rarely announces itself. It shows up as *smooth confidence*.\n\n**Early warning signs (meeting-saving edition):**\n- **Perfectly steady numbers** in a messy operation (real life has dents).\n- **Sudden step-change** with no operational explanation (new script, new hours, staffing change, system fix).\n- **High performance + low friction** (record results… but no one can explain the work).\n- **Disposition pile-ups:** too many items in “Other,” “No answer,” or “Left voicemail.”\n\n**Quick checks you can run in 10 minutes:**\n1) **Outlier scan:** Who is improbably best? Ask: “What would have to be true for this to be real?”\n2) **Lag test:** Do leading indicators move *before* outcomes? If outcomes move first, something’s backwards.\n3) **Sampling:** Pull 10 records. If 3 are weird, your dashboard is lying politely.\n\n**Dirty data rule of thumb:** if you can’t *sample it*, you can’t *trust it*. Dashboards don’t get offended; customers do.",{"id":116,"name":117,"type":88,"typeVersion":16,"position":118,"parameters":120,"category":94,"deletable":8,"connectable":8},"if_automation_vs_judgment","IF: Automation vs. judgment",[90,119],260,{"buttonId":71,"operator":93},{"id":122,"name":123,"type":98,"typeVersion":16,"position":124,"parameters":125,"category":50,"deletable":8,"connectable":8},"msg_automation_vs_judgment","When to trust automation",[100,119],{"text":126},"Automation is great at consistency. It’s terrible at context—unless you’ve paid for that context with good instrumentation.\n\n**Trust automation when:**\n- The decision is **reversible** (you can unwind it cheaply).\n- The input data is **instrumented and audited** (timestamps, sources, immutable logs).\n- The rule is **stable** (definition won’t change next month).\n- The cost of a false positive is low (e.g., send a reminder, not freeze an account).\n\n**Keep humans in the loop when:**\n- The decision is **high-stakes or reputational** (customer harm, compliance risk, escalation).\n- The data is **behavioral and squishy** (conversation summaries, sentiment, “intent”).\n- Incentives exist to game the metric (targets, rankings, branch comparisons).\n\n**Practical control:** start with automation as a **recommendation engine**, not a judge. Require a human sign-off until you’ve measured error rates with real samples (not vibes).",{"id":128,"name":129,"type":88,"typeVersion":16,"position":130,"parameters":132,"category":94,"deletable":8,"connectable":8},"if_messy_evidence","IF: Messy evidence → insight",[90,131],360,{"buttonId":74,"operator":93},{"id":134,"name":135,"type":98,"typeVersion":16,"position":136,"parameters":137,"category":50,"deletable":8,"connectable":8},"msg_messy_evidence","Use messy evidence without laundering it",[100,131],{"text":138},"Turning messy evidence into insight is not the same as “cleaning” it until it behaves.\n\n**Do this (truth-preserving):**\n- **Keep raw + curated side-by-side.** Curated views are for speed; raw is for disputes.\n- **Label uncertainty.** Add “confidence” or “unknown” explicitly—don’t bury it in averages.\n- **Keep contradictions.** If calls say one thing and outcomes say another, that’s not noise—that’s a lead.\n\n**Avoid this (truth-laundering):**\n- Throwing away “weird” records because they don’t fit the story.\n- Over-normalizing branch differences (you can normalize away the real operational constraint).\n- Converting narratives into a single score too early.\n\n**A good output format:**\n- 1–2 decision options\n- What would change our mind (the disconfirming test)\n- The top 3 assumptions that could be wrong\n\nIf you share the mess (what sources, what shape), I’ll suggest a decision-ready framing that keeps the sharp edges.",{"id":140,"name":141,"type":88,"typeVersion":16,"position":142,"parameters":144,"category":94,"deletable":8,"connectable":8},"if_compare_branches","IF: Comparing branches & attribution",[90,143],460,{"buttonId":77,"operator":93},{"id":146,"name":147,"type":98,"typeVersion":16,"position":148,"parameters":149,"category":50,"deletable":8,"connectable":8},"msg_compare_branches","Branch comparisons that don’t mislead",[100,143],{"text":150},"Branch comparisons fail for boring reasons—then ruin budgets for exciting reasons.\n\n**What teams repeatedly misread:**\n- **Different denominators:** one branch has more walk-ins; another has more appointments.\n- **Case-mix:** complex service needs make “speed” look bad and “quality” look invisible.\n- **Attribution gravity:** the last touch gets credit even when earlier work did the lifting.\n- **Operational constraints:** staffing, hours, local demand, nearby competitors.\n\n**Make comparisons decision-grade:**\n1) Compare **like-with-like** (segment by customer type, service, channel).\n2) Use **rates with credible denominators** (and show volume—rates lie at small N).\n3) Look for **within-branch change** before cross-branch ranking.\n4) Treat attribution as **a hypothesis**, not a fact—validate with small samples.\n\n**Meeting-saver line:** “Are we seeing performance—or are we seeing a different mix of work?”",{"id":152,"name":153,"type":88,"typeVersion":16,"position":154,"parameters":156,"category":94,"deletable":8,"connectable":8},"if_signal_culture","IF: Build a signal culture",[90,155],560,{"buttonId":80,"operator":93},{"id":158,"name":159,"type":98,"typeVersion":16,"position":160,"parameters":161,"category":50,"deletable":8,"connectable":8},"msg_signal_culture","Signal culture that drives decisions",[100,155],{"text":162},"A signal culture isn’t “more dashboards.” It’s *fewer surprises*.\n\n**Build it with these habits:**\n- **One metric, one owner, one definition.** If it changes, it gets versioned and announced.\n- **Sampling is standard.** Every key metric has a monthly (or weekly) spot-check.\n- **Disconfirming questions are rewarded.** The person who finds the flaw gets thanked, not side-eyed.\n- **Decision logs exist.** What we decided, based on what signals, and what we expected to happen.\n\n**Keep it human:**\n- Dashboards don’t make decisions—teams do. Your goal is to make the next decision easier, not the slide deck prettier.\n\nIf you tell me the decisions your team makes monthly, I’ll suggest the smallest set of signals worth maintaining (and the ones to retire).",{"id":164,"name":165,"type":88,"typeVersion":16,"position":166,"parameters":168,"category":94,"deletable":8,"connectable":8},"if_bring_human","IF: Bring in a human",[90,167],660,{"buttonId":83,"operator":93},{"id":170,"name":171,"type":172,"typeVersion":16,"position":173,"parameters":174,"category":178,"deletable":8,"connectable":8},"node_fallback","Handoff to human","fallback",[100,167],{"handoffMessage":175,"departmentId":176,"departmentName":177},"Got it—this sounds like a high-stakes call or a messy signal dispute. I’m handing this to a human reviewer. Share the metric name, source, time window, and one example record if you have it.","ops-analytics","Ops Analytics","terminal",[180,184,186,188,190,192,194,196,198,200,202,204,206,208,210,212],{"id":181,"source":35,"target":41,"sourceHandle":182,"targetHandle":183,"type":46},"conn_input_to_kb","out","in",{"id":185,"source":41,"target":52,"sourceHandle":182,"targetHandle":183,"type":46},"conn_kb_to_menu",{"id":187,"source":52,"target":86,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_trust",{"id":189,"source":52,"target":104,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_dirty",{"id":191,"source":52,"target":116,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_auto",{"id":193,"source":52,"target":128,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_messy",{"id":195,"source":52,"target":140,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_compare",{"id":197,"source":52,"target":152,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_culture",{"id":199,"source":52,"target":164,"sourceHandle":182,"targetHandle":183,"type":46},"conn_menu_to_if_human",{"id":201,"source":86,"target":96,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_trust_to_msg",{"id":203,"source":104,"target":110,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_dirty_to_msg",{"id":205,"source":116,"target":122,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_auto_to_msg",{"id":207,"source":128,"target":134,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_messy_to_msg",{"id":209,"source":140,"target":146,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_compare_to_msg",{"id":211,"source":152,"target":158,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_culture_to_msg",{"id":213,"source":164,"target":170,"sourceHandle":182,"targetHandle":183,"type":46},"conn_if_human_to_fallback","automation",[28,29,30,216,217,218],"metrics-trust","attribution","leadership-judgment",[220],"Calypso Messaging","intermediate","Calypso","2026-05-09T11:03:17.960Z","/en/workflows/branch-signal-decision-radar",{"en":224},{"title":9,"description":227,"ogDescription":228,"twitterDescription":229,"canonicalPath":224,"robots":230,"schemaType":231,"alternates":232},"Guide teams to trust the right branch signals, spot dirty data early, and decide when automation is safe—before confident mistakes happen.","A practical decision coach for branch ops: stress test metrics, catch dirty signal, avoid attribution traps, and know when humans must override automation.","Branch decision coach that helps you spot polished noise, test signal trust, and avoid attribution traps—plus guidance on when automation is safe.","index,follow","HowTo",[233],{"hreflang":6,"href":224},1778614429943]