[{"data":1,"prerenderedAt":194},["ShallowReactive",2],{"/en/workflows/signal-to-decision-coach-for-branch-teams":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":175,"tags":176,"integrations":177,"difficulty":180,"author":181,"verified":34,"featured":34,"date":182,"modified":182,"icon":7,"imageSrc":7,"path":183,"alternates":184,"seo":185},"cdd238cc-3ee6-41c9-87a9-701a861e239d","signal-to-decision-coach-for-branch-teams","en",null,true,"Signal-to-Decision Coach for Branch Teams","A guided chat workflow that helps teams test whether branch numbers are trustworthy, spot dirty signals early, and choose when to rely on automation vs human judgment.","## How it works\nThis workflow is a lightweight “decision safety check” you can run in chat before a meeting (or during one, when someone says a number with way too much confidence). It routes people to the right set of guardrails: which branch metrics deserve trust, how to detect dirty signal, when to trust automation, and how to turn messy evidence into usable insight without sanding off the inconvenient truth.\n\nIt’s built for the common failure mode: bad data that looks perfectly reasonable until it quietly steers a decision off a cliff. The workflow uses your Knowledge Base for consistent definitions and then guides users through decision-shaped checks with plain-language prompts.\n\n## Key features\n- Uses a Knowledge Base policy first to keep definitions, caveats, and metric “gotchas” consistent.\n- One-tap menu routes users to the exact decision check they need (numbers, dirty signal, automation vs judgment, culture).\n- Practical, meeting-ready advice designed to prevent confident wrong decisions—not to produce prettier slides.\n- Optional handoff to an Analytics Support department when the situation needs human review.\n\n## Step-by-step\n1. **Trigger:** A user starts the workflow.\n2. **Knowledge Base grounding:** The workflow enables Knowledge Base guidance so responses follow your approved metric definitions and caveats.\n3. **Decision menu:** The user picks what they’re trying to decide:\n   - **Trust this branch number?**\n   - **Spot dirty signal fast**\n   - **Automation vs judgment**\n   - **Build a signal culture**\n   - **Talk to an analyst**\n4. **Routed guidance:** Based on the button clicked, the workflow sends a targeted checklist-style message with practical warning signs and next steps.\n5. **Escalation (optional):** If the user chooses **Talk to an analyst**, the workflow hands the conversation to **Analytics Support**.\n\n## Setup requirements\n- **Calypso Knowledge Base**: Recommended (this workflow turns on Knowledge Base guidance). No additional credentials are required inside this template.\n- **Routing setup**: Create/confirm an **Analytics Support** department (or update the department ID/name in the handoff step).",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":149,"routingEnabled":8,"active":34},"wf_signal_to_decision_coach_v1","calypso-public-library","1.0.0",1,[18,35,41,53,81,90,96,102,107,113,121,127,133,139],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":24,"category":33,"deletable":34,"connectable":34},"node_flow_configs","Workflow settings","flow-configs",[23,23],80,{"name":9,"description":25,"tags":26,"triggerType":32},"Guides users through decision-shaped checks to validate branch numbers, spot dirty signal, and decide when automation needs human judgment.",[27,28,29,30,31],"signal-design","decision-systems","branch-metrics","data-quality","attribution","input","policy",false,{"id":36,"name":37,"type":32,"typeVersion":16,"position":38,"parameters":40,"category":32,"deletable":34,"connectable":8},"node_input","Start",[23,39],200,{},{"id":42,"name":43,"type":44,"typeVersion":16,"position":45,"parameters":47,"category":52,"deletable":8,"connectable":8},"node_kb_policy","Knowledge base grounding","knowledge-base-policy",[46,39],320,{"enabled":8,"fallbackToRouting":8,"sticky":8,"stickyMode":48,"activationOpener":49,"personalization":51},"default",{"enabled":8,"instruction":50},"Use the Knowledge Base to keep metric definitions, attribution rules, and branch reporting caveats consistent. Prefer practical, decision-ready guidance over theory. If definitions are ambiguous, ask for the metric name, date range, and source before advising.",{"useContactName":34},"response",{"id":54,"name":55,"type":56,"typeVersion":16,"position":57,"parameters":59,"category":52,"deletable":8,"connectable":8},"node_menu","Pick a decision check","interactive-message",[58,39],560,{"messageType":60,"headerText":61,"bodyText":62,"footerText":63,"sectionTitle":64,"buttons":65,"ctaDisplayText":64,"ctaUrl":64},"button","Decision check (2 minutes)","Pick what you’re trying to decide. I’ll help you separate trustworthy signal from polished noise—before a confident meeting makes a confident wrong call.","Tip: No source + timeframe? Not decision-ready.","",[66,69,72,75,78],{"id":67,"title":68},"check_numbers","Trust this number?",{"id":70,"title":71},"spot_dirty","Spot dirty signal",{"id":73,"title":74},"auto_vs_human","Auto vs judgment",{"id":76,"title":77},"culture","Signal culture",{"id":79,"title":80},"handoff_analyst","Talk to an analyst",{"id":82,"name":83,"type":84,"typeVersion":16,"position":85,"parameters":87,"category":89,"deletable":8,"connectable":8},"node_if_check_numbers","If: Trust this branch number","if",[86,23],820,{"buttonId":67,"operator":88},"equals","routing",{"id":91,"name":92,"type":84,"typeVersion":16,"position":93,"parameters":95,"category":89,"deletable":8,"connectable":8},"node_if_spot_dirty","If: Spot dirty signal fast",[86,94],160,{"buttonId":70,"operator":88},{"id":97,"name":98,"type":84,"typeVersion":16,"position":99,"parameters":101,"category":89,"deletable":8,"connectable":8},"node_if_auto_vs_human","If: Automation vs judgment",[86,100],240,{"buttonId":73,"operator":88},{"id":103,"name":104,"type":84,"typeVersion":16,"position":105,"parameters":106,"category":89,"deletable":8,"connectable":8},"node_if_culture","If: Build a signal culture",[86,46],{"buttonId":76,"operator":88},{"id":108,"name":109,"type":84,"typeVersion":16,"position":110,"parameters":112,"category":89,"deletable":8,"connectable":8},"node_if_handoff","If: Talk to an analyst",[86,111],400,{"buttonId":79,"operator":88},{"id":114,"name":115,"type":116,"typeVersion":16,"position":117,"parameters":119,"category":52,"deletable":8,"connectable":8},"node_txt_check_numbers","Guidance: Which numbers to trust","text-message",[118,23],1080,{"text":120},"Here’s the fast way to tell if a branch number deserves trust—or just applause:\n\n1) **Definition check (non‑negotiable):** What exactly counts? If two people can describe it differently, the trend is decorative.\n2) **Denominator sanity:** Are we comparing per‑branch, per‑customer, per‑staff hour, or raw totals? Raw totals love large branches and busy days.\n3) **Coverage + missingness:** Ask “What % of events could this metric even see?” Blind spots create fake winners.\n4) **Lag + backfill:** Does it change after the fact? If yes, don’t treat yesterday’s number like a verdict.\n5) **Branch comparability:** Any policy changes, staffing shifts, local promos, or channel mix differences? If branches aren’t playing the same game, the leaderboard lies.\n\nRule of thumb: **If the metric can’t survive a 60‑second explanation of source + definition + timeframe, it shouldn’t survive a 60‑minute meeting.**",{"id":122,"name":123,"type":116,"typeVersion":16,"position":124,"parameters":125,"category":52,"deletable":8,"connectable":8},"node_txt_spot_dirty","Guidance: Spot dirty signal",[118,94],{"text":126},"Dirty signal usually looks clean until you ask one irritating question. Use these tripwires:\n\n- **Too-smooth trend:** Real operations are lumpy. Perfectly smooth lines often mean filtering, averaging, or delayed logging.\n- **Step-change with no story:** If the chart jumps but the branch can’t name what changed, suspect instrumentation or definition drift.\n- **One source says up, another says flat:** Mismatched sources aren’t “different perspectives”—they’re a decision hazard.\n- **Survivorship:** Are we only counting completed conversations/visits/tickets? Excluding drop-offs makes everything look great.\n- **Attribution magnetism:** If every initiative ‘drives’ the same outcome, your attribution is probably double-counting.\n\nMeeting-saving move: ask **“What would make this number wrong?”** If the room goes quiet, you just found the work.",{"id":128,"name":129,"type":116,"typeVersion":16,"position":130,"parameters":131,"category":52,"deletable":8,"connectable":8},"node_txt_auto_vs_human","Guidance: Automation vs judgment",[118,100],{"text":132},"When to trust automation vs when you still need human judgment:\n\n**Automation is great when:**\n- The definition is stable and everyone agrees on it.\n- Inputs are complete enough (coverage is known, not assumed).\n- Errors are **bounded** (a wrong answer is annoying, not catastrophic).\n- You can monitor drift (the world changes; dashboards rarely complain).\n\n**Humans must stay in the loop when:**\n- The decision is high-stakes (staffing, compliance, customer harm).\n- Data is sparse, newly instrumented, or heavily backfilled.\n- Incentives are active (people will optimize the metric, not the outcome).\n- Branch context matters (local events, outages, policy exceptions).\n\nPractical policy: **Automate recommendations, not consequences.** Let systems suggest; require humans to approve when the cost of being wrong is asymmetric.",{"id":134,"name":135,"type":116,"typeVersion":16,"position":136,"parameters":137,"category":52,"deletable":8,"connectable":8},"node_txt_culture","Guidance: Build signal culture",[118,46],{"text":138},"A strong signal culture speeds decisions because it reduces argument-by-slide. A few habits that actually work:\n\n- **Every metric gets a ‘label’:** owner, definition, source, refresh cadence, known blind spots.\n- **Pre‑mortem the metric:** write down the top 3 ways it can lie (missingness, selection bias, attribution overlap).\n- **Decision > dashboard:** require a sentence like “If this goes up/down, we will do X.” If no action, it’s trivia.\n- **Compare branches responsibly:** normalize (per customer / per hour), note policy/channel mix, and call out measurement differences.\n- **Reward truth-telling:** celebrate ‘we found the data was wrong’ as progress, not embarrassment.\n\nWitty but true: **the first casualty of a KPI is usually curiosity. Guard it.**",{"id":140,"name":141,"type":142,"typeVersion":16,"position":143,"parameters":144,"category":148,"deletable":8,"connectable":8},"node_fallback_handoff","Handoff: Analytics Support","fallback",[118,111],{"handoffMessage":145,"departmentId":146,"departmentName":147},"Got it. I’m handing this to Analytics Support so a human can review the metric definition, source data, and any branch-level comparability issues. If you can, share: metric name, timeframe, branches compared, and where the number came from.","analytics-support","Analytics Support","terminal",[150,153,155,157,159,161,163,165,167,169,171,173],{"id":151,"source":36,"target":42,"sourceHandle":152,"targetHandle":152,"type":48},"conn_input_to_kb","main",{"id":154,"source":42,"target":54,"sourceHandle":152,"targetHandle":152,"type":48},"conn_kb_to_menu",{"id":156,"source":54,"target":82,"sourceHandle":152,"targetHandle":152,"type":48},"conn_menu_to_if_numbers",{"id":158,"source":54,"target":91,"sourceHandle":152,"targetHandle":152,"type":48},"conn_menu_to_if_dirty",{"id":160,"source":54,"target":97,"sourceHandle":152,"targetHandle":152,"type":48},"conn_menu_to_if_auto",{"id":162,"source":54,"target":103,"sourceHandle":152,"targetHandle":152,"type":48},"conn_menu_to_if_culture",{"id":164,"source":54,"target":108,"sourceHandle":152,"targetHandle":152,"type":48},"conn_menu_to_if_handoff",{"id":166,"source":82,"target":114,"sourceHandle":152,"targetHandle":152,"type":48},"conn_if_numbers_to_txt",{"id":168,"source":91,"target":122,"sourceHandle":152,"targetHandle":152,"type":48},"conn_if_dirty_to_txt",{"id":170,"source":97,"target":128,"sourceHandle":152,"targetHandle":152,"type":48},"conn_if_auto_to_txt",{"id":172,"source":103,"target":134,"sourceHandle":152,"targetHandle":152,"type":48},"conn_if_culture_to_txt",{"id":174,"source":108,"target":140,"sourceHandle":152,"targetHandle":152,"type":48},"conn_if_handoff_to_fallback","automation",[27,28,29,30,31],[178,179],"Calypso Knowledge Base","Web chat","intermediate","Calypso","2026-06-16T11:03:36.861Z","/en/workflows/signal-to-decision-coach-for-branch-teams",{"en":183},{"title":186,"description":187,"ogDescription":188,"twitterDescription":189,"canonicalPath":183,"robots":190,"schemaType":191,"alternates":192},"Signal to Decision Coach for Branch Teams","Guide branch teams to trust the right metrics, spot dirty signals early, and know when automation needs human judgment.","A chat based decision coach for branch metrics: reality check numbers, catch dirty signal before meetings, and escalate to Analytics Support when needed.","Help teams trust the right branch numbers, spot dirty signals early, and decide when automation needs human judgment—without the academic fluff.","index,follow","HowTo",[193],{"hreflang":6,"href":183},1785947669140]