[{"data":1,"prerenderedAt":283},["ShallowReactive",2],{"/en/workflows/decision-signals-triage-for-branch-leaders":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":264,"tags":265,"integrations":267,"difficulty":270,"author":271,"verified":34,"featured":34,"date":272,"modified":272,"icon":7,"imageSrc":7,"path":273,"alternates":274,"seo":275},"73d66193-930d-4d07-8cbf-e9553ee081c8","decision-signals-triage-for-branch-leaders","en",null,true,"Decision Signals Triage for Branch Leaders","A guided menu that helps teams separate trustworthy branch signals from polished noise, spot dirty data before meetings, and decide when automation needs human judgment.","## How it works\nThis workflow turns messy branch metrics, conversation notes, and “looks-good” dashboards into decision-ready signal checks—without turning your team into a statistics club. It starts with a Knowledge Base–powered coach, then routes the user through a practical menu of decision-shaped topics.\n\nUse it when a meeting is headed toward a confident wrong call: the numbers are clean, the story is neat, and the risk is hidden. The workflow prompts the user to stress-test what they’re about to trust, what to measure next, and where humans still need to intervene.\n\n## Key features\n- Knowledge Base policy enabled up front for consistent, grounded answers before routing\n- Interactive menu covering six high-leverage decision failure modes (trust, dirt, automation, messy evidence, comparisons, culture)\n- Short, operator-friendly guidance designed to be used *right before* a decision meeting\n- “Back to menu” loop to keep the conversation moving without restarting\n- One-click handoff to a human team via fallback routing when judgment is required\n\n## Step-by-step\n1. **Trigger:** The workflow starts when a user opens the flow (Input).\n2. **Apply Knowledge Base policy:** Enables KB-guided responses and keeps routing available.\n3. **Set expectations:** A short intro explains the goal: avoid confident wrong decisions by stress-testing signals.\n4. **Choose a signal triage path:** The user selects one of six options (Interactive Message).\n5. **Route by selection:** IF nodes match the chosen button and send the relevant guidance as a text message.\n6. **Choose what to do next:** The user can return to the main menu or request a human handoff.\n7. **Handoff (optional):** If requested, the workflow routes to a fallback handoff for a human review.\n\n## Setup requirements\n- A Calypso Knowledge Base connected to your organization’s signal definitions (e.g., branch KPI glossary, attribution rules, CRM logging standards).\n- Optional: A named department (e.g., “Analytics Ops”) configured to receive handoffs.\n- No external credentials are required by this workflow as built.",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":212,"routingEnabled":8,"active":34},"wf_decision_signals_triage_v1","calypso-public-library","1.0.0",1,[18,35,42,54,62,93,102,108,113,119,125,131,138,144,150,156,162,168,182,189,195,202],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":33,"deletable":34,"connectable":34},"node_flow_cfg","Flow settings","flow-configs",[23,24],-80,-40,{"name":9,"description":26,"tags":27,"triggerType":32},"Guided triage to separate trustworthy branch signals from polished noise, spot dirty data before meetings, and decide when automation needs human judgment.",[28,29,30,31],"signal-design","data-quality","branch-metrics","decision-systems","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],100,120,{},{"id":43,"name":44,"type":45,"typeVersion":16,"position":46,"parameters":48,"category":53,"deletable":8,"connectable":8},"node_kb_policy","Knowledge base coach","knowledge-base-policy",[47,40],320,{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":49,"activationOpener":50,"personalization":52},"default",{"enabled":8,"instruction":51},"Be practical and direct. Treat clean-looking data as suspicious until proven otherwise. Avoid academic phrasing; use short checklists and decision prompts.",{"useContactName":34},"response",{"id":55,"name":56,"type":57,"typeVersion":16,"position":58,"parameters":60,"category":53,"deletable":8,"connectable":8},"node_intro","Set the frame","text-message",[59,40],560,{"text":61},"This is a quick signal triage—meant for the 10 minutes before someone says “the dashboard is clear.”\n\nPick what you’re deciding, then we’ll stress-test the signal *just enough* to avoid a confident wrong turn.",{"id":63,"name":64,"type":65,"typeVersion":16,"position":66,"parameters":68,"category":53,"deletable":8,"connectable":8},"node_menu","Pick a signal check","interactive-message",[67,40],820,{"messageType":69,"headerText":70,"bodyText":71,"footerText":72,"sectionTitle":73,"buttons":74,"ctaDisplayText":73,"ctaUrl":73},"button","Decision Signals Triage","Where is the decision most likely to go wrong? Choose the check that matches your situation.","If it looks perfect, start with Trust vs noise.","",[75,78,81,84,87,90],{"id":76,"title":77},"polished_noise_vs_trust","Trust vs noise",{"id":79,"title":80},"spot_dirty_signal","Dirty signal",{"id":82,"title":83},"automation_vs_judgment","Auto vs human",{"id":85,"title":86},"messy_evidence_to_insight","Messy evidence",{"id":88,"title":89},"comparison_misreads","Compare branches",{"id":91,"title":92},"signal_culture","Signal culture",{"id":94,"name":95,"type":96,"typeVersion":16,"position":97,"parameters":99,"category":101,"deletable":8,"connectable":8},"node_if_polished","If: polished noise vs trust","if",[98,24],1060,{"buttonId":76,"operator":100},"equals","routing",{"id":103,"name":104,"type":96,"typeVersion":16,"position":105,"parameters":107,"category":101,"deletable":8,"connectable":8},"node_if_dirty","If: spot dirty signal",[98,106],40,{"buttonId":79,"operator":100},{"id":109,"name":110,"type":96,"typeVersion":16,"position":111,"parameters":112,"category":101,"deletable":8,"connectable":8},"node_if_auto","If: automation vs judgment",[98,40],{"buttonId":82,"operator":100},{"id":114,"name":115,"type":96,"typeVersion":16,"position":116,"parameters":118,"category":101,"deletable":8,"connectable":8},"node_if_messy","If: messy evidence to insight",[98,117],200,{"buttonId":85,"operator":100},{"id":120,"name":121,"type":96,"typeVersion":16,"position":122,"parameters":124,"category":101,"deletable":8,"connectable":8},"node_if_compare","If: comparison misreads",[98,123],280,{"buttonId":88,"operator":100},{"id":126,"name":127,"type":96,"typeVersion":16,"position":128,"parameters":130,"category":101,"deletable":8,"connectable":8},"node_if_culture","If: signal culture",[98,129],360,{"buttonId":91,"operator":100},{"id":132,"name":133,"type":57,"typeVersion":16,"position":134,"parameters":136,"category":53,"deletable":8,"connectable":8},"node_text_polished","Trustworthy numbers vs polished noise",[135,24],1300,{"text":137},"**Trustworthy branch numbers usually have scars.** Polished noise looks smooth.\n\nQuick trust checks:\n- **Traceability:** Can you point to *where* each number comes from (system + timestamp + definition), not just the chart?\n- **Friction signals:** Do you see expected “mess” (returns, reversals, late logs), or is it suspiciously tidy?\n- **Denominator sanity:** If a rate moved, did the underlying volume change too? (Rates lie best when volumes shrink.)\n- **Cross-source agreement:** Does it roughly align with at least one independent source (e.g., POS vs CRM)?\n\nWhat’s usually noise:\n- Rankings without confidence bands (“Branch 12 is #1!”) when sample sizes differ.\n- Composite scores nobody can explain in one breath.\n\nDecision move: **Name the decision, then name the one number that could embarrass it next week. Go validate that number first.**",{"id":139,"name":140,"type":57,"typeVersion":16,"position":141,"parameters":142,"category":53,"deletable":8,"connectable":8},"node_text_dirty","Spot dirty signal before the meeting",[135,106],{"text":143},"Dirty signal rarely shows up as “wrong.” It shows up as **too consistent**.\n\nFast pre-meeting checks (10 minutes, no heroics):\n1) **Freshness:** Any sudden “flatline” after a system change, holiday, or staffing shift?\n2) **Outliers with stories:** Find the top/bottom 3 branches—can anyone explain them without hand-waving?\n3) **Missingness drift:** Did “unknown/other” buckets grow? That’s your silent data leak.\n4) **Definition drift:** Did the metric definition quietly change (new funnel step, new channel tagging)?\n\nMeeting-saver line: **“Before we argue, what would we expect to see if the data were broken?”**\nIf the room can’t answer, the signal isn’t meeting-ready.",{"id":145,"name":146,"type":57,"typeVersion":16,"position":147,"parameters":148,"category":53,"deletable":8,"connectable":8},"node_text_auto","Automation vs human judgment",[135,40],{"text":149},"Automation is great at **repeatable checks**. Humans are better at **context, exceptions, and ethics**.\n\nTrust automation when:\n- The rule is stable (same inputs → same action), and the cost of a miss is low.\n- You can monitor it with a simple tripwire (e.g., volume drops, distribution shifts).\n\nInsist on human judgment when:\n- **Incentives are involved** (people will optimize the metric, not the mission).\n- The data can be gamed, delayed, or “cleaned” by process changes.\n- The action is hard to reverse (staffing cuts, branch closures, major budget moves).\n\nPractical guardrail: **Automate the recommendation, not the decision.** Require a human to answer: “What would change my mind?”",{"id":151,"name":152,"type":57,"typeVersion":16,"position":153,"parameters":154,"category":53,"deletable":8,"connectable":8},"node_text_messy","Messy evidence to usable insight",[135,117],{"text":155},"The goal isn’t to make evidence spotless—it’s to make it **decision-usable without sanding off the truth**.\n\nHow to keep the signal while handling mess:\n- **Keep the raw story nearby:** Pair the metric with 3 real examples (calls, notes, incidents). Numbers without examples become fiction.\n- **Separate “unknown” from “zero”:** Unknown is a data problem; zero is a business outcome.\n- **Use ranges, not single points:** “Likely 8–12%” beats “10.3%” when logs are imperfect.\n- **Annotate reality:** Staffing changes, promotions, outages—write them next to the trend so you don’t blame the wrong cause.\n\nDecision move: **Make one ‘good enough’ call now, and write the one data improvement that would make the next call easier.**",{"id":157,"name":158,"type":57,"typeVersion":16,"position":159,"parameters":160,"category":53,"deletable":8,"connectable":8},"node_text_compare","Comparing branches without fooling yourself",[135,123],{"text":161},"Teams misread branch comparisons in very predictable ways—because dashboards reward confidence.\n\nCommon comparison traps:\n- **Different mixes:** Branches serve different customer types; comparing raw conversion is often comparing *mix*, not performance.\n- **Small samples:** The loudest “winner” is often just variance wearing a suit.\n- **Attribution artifacts:** If one branch logs better, it “wins” more—regardless of actual outcomes.\n- **Process differences:** One branch counts leads differently, closes tickets faster, or tags channels more aggressively.\n\nSafer comparison moves:\n- Compare **within segments** (same product, same channel, similar volume).\n- Look for **consistency over time**, not one-week hero charts.\n\nDecision move: **Ask ‘What’s the fairest comparison?’ before asking ‘Who’s best?’**",{"id":163,"name":164,"type":57,"typeVersion":16,"position":165,"parameters":166,"category":53,"deletable":8,"connectable":8},"node_text_culture","Build a signal culture that produces decisions",[135,129],{"text":167},"A healthy signal culture is not “more metrics.” It’s **fewer arguments and faster, reversible decisions**.\n\nWhat helps (and what usually fails first):\n- **Single source of definitions:** If people debate definitions in meetings, you don’t have metrics—you have opinions with charts.\n- **Pre-mortems:** “How could this number be misleading?” before “What should we do?”\n- **Visible uncertainty:** Encourage “I’m 70% sure” language. False certainty is the first lie.\n- **Decision logs:** Write what you decided, what you trusted, and what would trigger a revisit.\n\nWit-sized truth: **If a metric can’t survive one skeptical question, it doesn’t deserve to steer a branch.**",{"id":169,"name":170,"type":65,"typeVersion":16,"position":171,"parameters":173,"category":53,"deletable":8,"connectable":8},"node_next","Next step",[172,40],1540,{"messageType":69,"headerText":170,"bodyText":174,"footerText":73,"sectionTitle":73,"buttons":175,"ctaDisplayText":73,"ctaUrl":73},"Want another check, or should this go to a human for review?",[176,179],{"id":177,"title":178},"back_to_menu","Another check",{"id":180,"title":181},"handoff_human","Human review",{"id":183,"name":184,"type":96,"typeVersion":16,"position":185,"parameters":188,"category":101,"deletable":8,"connectable":8},"node_if_back","If: another check",[186,187],1780,80,{"buttonId":177,"operator":100},{"id":190,"name":191,"type":96,"typeVersion":16,"position":192,"parameters":194,"category":101,"deletable":8,"connectable":8},"node_if_handoff","If: human review",[186,193],160,{"buttonId":180,"operator":100},{"id":196,"name":197,"type":57,"typeVersion":16,"position":198,"parameters":200,"category":53,"deletable":8,"connectable":8},"node_text_restart","Run another check",[199,187],2020,{"text":201},"To run another check, reopen **Decision Signals Triage** and pick a different option.\n\nSmall habit that saves big meetings: pick the *one* check most likely to overturn today’s decision, do it, then come back for the second.",{"id":203,"name":204,"type":205,"typeVersion":16,"position":206,"parameters":207,"category":211,"deletable":8,"connectable":8},"node_fallback","Handoff: Analytics Ops","fallback",[199,193],{"handoffMessage":208,"departmentId":209,"departmentName":210},"Got it—routing this to Analytics Ops for a human signal review. Please share: the decision you’re making, the metric(s) you’re using, and what would change your mind.","dept_analytics_ops","Analytics Ops","terminal",[213,215,217,219,221,223,225,227,229,231,234,236,238,240,242,244,246,248,250,252,254,256,258,260,262],{"id":214,"source":36,"target":43,"sourceHandle":49,"targetHandle":49,"type":49},"conn_input_to_kb",{"id":216,"source":43,"target":55,"sourceHandle":49,"targetHandle":49,"type":49},"conn_kb_to_intro",{"id":218,"source":55,"target":63,"sourceHandle":49,"targetHandle":49,"type":49},"conn_intro_to_menu",{"id":220,"source":63,"target":94,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_polished",{"id":222,"source":63,"target":103,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_dirty",{"id":224,"source":63,"target":109,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_auto",{"id":226,"source":63,"target":114,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_messy",{"id":228,"source":63,"target":120,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_compare",{"id":230,"source":63,"target":126,"sourceHandle":49,"targetHandle":49,"type":49},"conn_menu_to_if_culture",{"id":232,"source":94,"target":132,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_polished_to_text","true",{"id":235,"source":103,"target":139,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_dirty_to_text",{"id":237,"source":109,"target":145,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_auto_to_text",{"id":239,"source":114,"target":151,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_messy_to_text",{"id":241,"source":120,"target":157,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_compare_to_text",{"id":243,"source":126,"target":163,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_culture_to_text",{"id":245,"source":132,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_polished_to_next",{"id":247,"source":139,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_dirty_to_next",{"id":249,"source":145,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_auto_to_next",{"id":251,"source":151,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_messy_to_next",{"id":253,"source":157,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_compare_to_next",{"id":255,"source":163,"target":169,"sourceHandle":49,"targetHandle":49,"type":49},"conn_text_culture_to_next",{"id":257,"source":169,"target":183,"sourceHandle":49,"targetHandle":49,"type":49},"conn_next_to_if_back",{"id":259,"source":169,"target":190,"sourceHandle":49,"targetHandle":49,"type":49},"conn_next_to_if_handoff",{"id":261,"source":183,"target":196,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_back_to_restart_text",{"id":263,"source":190,"target":203,"sourceHandle":233,"targetHandle":49,"type":49},"conn_if_handoff_to_fallback","automation",[28,29,30,31,266],"research-triage",[268,269],"Calypso Knowledge Base","Calypso Inbox","intermediate","Calypso","2026-05-04T11:03:30.814Z","/en/workflows/decision-signals-triage-for-branch-leaders",{"en":273},{"title":9,"description":276,"ogDescription":277,"twitterDescription":278,"canonicalPath":273,"robots":279,"schemaType":280,"alternates":281},"Route teams through practical signal checks so branch decisions use trustworthy numbers, catch dirty data, and know when humans must step in.","A guided signal triage menu for branch leaders: trust the right numbers, catch dirty signals early, and choose when automation needs human judgment.","Turn messy branch evidence into decision ready signal checks. Spot polished noise, detect dirty data, and know when to escalate to humans.","index,follow","HowTo",[282],{"hreflang":6,"href":273},1778614431187]