[{"data":1,"prerenderedAt":483},["ShallowReactive",2],{"/en/workflows/signal-reliability-playbook-for-branch-decisions":3},{"id":4,"slug":5,"locale":6,"translationGroupId":7,"localeSwitchApproved":8,"title":9,"description":10,"documentationMarkdown":11,"workflowJson":12,"category":462,"tags":463,"integrations":467,"difficulty":470,"author":471,"verified":33,"featured":33,"date":472,"modified":472,"icon":7,"imageSrc":7,"path":473,"alternates":474,"seo":475},"832c99d6-97f2-4771-a913-a3a573ba6d48","signal-reliability-playbook-for-branch-decisions","en",null,true,"Signal Reliability Playbook for Branch Decisions","A guided chat workflow that helps teams sanity-check branch numbers, spot dirty signals early, and choose when to trust automation vs human judgment—before decisions calcify.","## How it works\nThis workflow is a decision-focused chat guide for operators and leaders who need to turn messy branch activity, conversations, and attribution into actions they won’t regret in the next weekly meeting. It starts with a knowledge-base assist (so repeat questions don’t become repeat debates), then routes the user into a set of practical “truth tests” for common signal failures.\n\nIt’s designed for the reality that bad data often looks polished—right up until someone presents it with confidence. Each path gives short, usable checks: what to trust, what to challenge, what to measure next, and when to escalate to a human analyst.\n\n## Key features\n- Knowledge-base response policy first, then guided routing when the answer needs judgment\n- Button-driven menu for decision-shaped topics (trustworthy numbers, dirty signals, automation vs human review, attribution misreads)\n- Consistent next-step actions after every answer: go back to the menu or hand off to an analyst\n- “Unknown selection” guardrails that bring users back to the menu instead of dead-ending the conversation\n\n## Step-by-step\n1. **Trigger:** A user starts the workflow (**Input**).\n2. **Knowledge assist:** The workflow applies the **Knowledge Base Policy** to answer straightforward questions; if it can’t, it continues to routing.\n3. **Choose a decision checkpoint:** The user sees the **Signal checkpoints** menu (buttons).\n4. **Routing:** The workflow matches the selected button and sends a targeted guidance message:\n   1. **Which numbers deserve trust (vs polished noise)**\n   2. **How to spot dirty signal before a meeting goes off the rails**\n   3. **When to trust automation vs require human judgment**\n   4. **How to use messy evidence without “cleaning away” the truth**\n   5. **What teams misread when comparing branches, conversations, and attribution**\n   6. **How to build a signal culture that produces decisions, not slides**\n   7. **Talk to an analyst** (handoff)\n5. **Next step:** After any guidance message, the user gets **What would you like to do next?**\n   - **Talk to an analyst** → routed to **Fallback (handoff)**\n   - **Back to checkpoints** → returns to the main menu\n6. **If the user doesn’t pick a button:** The workflow sends a short correction and returns them to the main menu.\n\n## Setup requirements\n- A Calypso knowledge base connected to this workflow (recommended for best results).\n- No additional credentials are required for the workflow itself.\n- If you want live escalation, configure a receiving team/department for the handoff (e.g., “Data & Insights”).",{"id":13,"teamId":14,"name":9,"version":15,"workflowVersion":16,"nodes":17,"connections":348,"routingEnabled":8,"active":33},"wf_signal_reliability_playbook_branch_decisions","calypso-public-library","1.0.1",1,[18,34,40,52,85,95,100,105,111,117,123,128,136,142,148,154,160,166,180,187,193,214,220,225,230,235,240,245,250,256,261,266,271,276,281,294,300,305,312,319,325,331,338],{"id":19,"name":20,"type":21,"typeVersion":16,"position":22,"parameters":25,"category":32,"deletable":33,"connectable":33},"node_flow_cfg","Flow settings","flow-configs",[23,24],-240,120,{"name":9,"description":26,"tags":27,"triggerType":31},"Guided checkpoints to test branch signal reliability, spot dirty data, and choose when automation needs human judgment.",[28,29,30],"signal-quality","decision-making","branch-metrics","input","policy",false,{"id":35,"name":36,"type":31,"typeVersion":16,"position":37,"parameters":39,"category":31,"deletable":33,"connectable":8},"node_input","Input",[38,24],-40,{},{"id":41,"name":42,"type":43,"typeVersion":16,"position":44,"parameters":46,"category":51,"deletable":8,"connectable":8},"node_kb","Knowledge base assist","knowledge-base-policy",[45,24],180,{"enabled":8,"fallbackToRouting":8,"sticky":33,"stickyMode":47,"activationOpener":48,"personalization":50},"default",{"enabled":8,"instruction":49},"Use the knowledge base for clear, factual questions. If the user is asking for judgment (what to trust, what to measure, what might be wrong), route them into the checkpoints menu.",{"useContactName":33},"response",{"id":53,"name":54,"type":55,"typeVersion":16,"position":56,"parameters":58,"category":51,"deletable":8,"connectable":8},"node_menu","Signal checkpoints","interactive-message",[57,24],420,{"messageType":59,"headerText":54,"bodyText":60,"footerText":61,"sectionTitle":62,"buttons":63,"ctaDisplayText":62,"ctaUrl":62},"button","Pick what you’re trying to decide. I’ll help you separate reliable signals from polished noise—before the confident meeting makes it permanent.","Tip: Choose one checkpoint, then you can pick one more.","",[64,67,70,73,76,79,82],{"id":65,"title":66},"btn_trust_numbers","Trust numbers",{"id":68,"title":69},"btn_dirty_signal","Dirty signal",{"id":71,"title":72},"btn_auto_vs_human","Auto vs judgment",{"id":74,"title":75},"btn_messy_evidence","Messy evidence",{"id":77,"title":78},"btn_compare_misreads","Branch comparison",{"id":80,"title":81},"btn_signal_culture","Signal culture",{"id":83,"title":84},"btn_talk_to_analyst","Talk to analyst",{"id":86,"name":87,"type":88,"typeVersion":16,"position":89,"parameters":92,"category":94,"deletable":8,"connectable":8},"node_if_trust","If: trust numbers","if",[90,91],660,60,{"buttonId":65,"operator":93},"equals","routing",{"id":96,"name":97,"type":88,"typeVersion":16,"position":98,"parameters":99,"category":94,"deletable":8,"connectable":8},"node_if_dirty","If: dirty signal",[90,24],{"buttonId":68,"operator":93},{"id":101,"name":102,"type":88,"typeVersion":16,"position":103,"parameters":104,"category":94,"deletable":8,"connectable":8},"node_if_auto","If: automation vs judgment",[90,45],{"buttonId":71,"operator":93},{"id":106,"name":107,"type":88,"typeVersion":16,"position":108,"parameters":110,"category":94,"deletable":8,"connectable":8},"node_if_messy","If: messy evidence",[90,109],240,{"buttonId":74,"operator":93},{"id":112,"name":113,"type":88,"typeVersion":16,"position":114,"parameters":116,"category":94,"deletable":8,"connectable":8},"node_if_compare","If: compare misreads",[90,115],300,{"buttonId":77,"operator":93},{"id":118,"name":119,"type":88,"typeVersion":16,"position":120,"parameters":122,"category":94,"deletable":8,"connectable":8},"node_if_culture","If: signal culture",[90,121],360,{"buttonId":80,"operator":93},{"id":124,"name":125,"type":88,"typeVersion":16,"position":126,"parameters":127,"category":94,"deletable":8,"connectable":8},"node_if_talk","If: talk to analyst",[90,57],{"buttonId":83,"operator":93},{"id":129,"name":130,"type":131,"typeVersion":16,"position":132,"parameters":134,"category":51,"deletable":8,"connectable":8},"node_text_trust","Guidance: trust numbers","text-message",[133,91],940,{"text":135},"**Trustworthy branch numbers usually have three things: definition, coverage, and friction.**\n\n1) **Definition:** Can two people describe the metric the same way? If not, it’s a vibe wearing a spreadsheet.\n2) **Coverage:** Does it include the boring cases (no-shows, cancellations, reopens), or only the clean wins?\n3) **Friction:** Reliable numbers are a little annoying to move. If a metric jumps 25% with zero operational change, it’s often a tracking change, not reality.\n\n**Fast tests:**\n- Ask “What would make this number go down even if we did a great job?” If nobody can answer, it’s probably self-congratulatory.\n- Compare **rate + volume** (e.g., conversion rate *and* leads). Rates alone can flatter you.\n- Look for the **audit trail**: timestamps, sources, and consistent counting rules.\n\nIf you only trust one thing this week: trust the metric that can survive one skeptical question without a footnote explosion.",{"id":137,"name":138,"type":131,"typeVersion":16,"position":139,"parameters":140,"category":51,"deletable":8,"connectable":8},"node_text_dirty","Guidance: dirty signal",[133,24],{"text":141},"**Dirty signal isn’t loud. It’s polite.** It shows up as numbers that look consistent while the real world feels inconsistent.\n\n**Early warning signs (meeting-proof):**\n- **Too-smooth trends:** real operations are lumpy. Perfect diagonals often mean smoothing, batching, or backfills.\n- **Timestamp weirdness:** spikes at midnight, end-of-week cliffs, or sudden shifts after a system change.\n- **One field does all the work:** if attribution depends on a single tag/UTM/campaign field, assume it’s wrong in the most convenient way.\n- **Mismatch across sources:** branch events say one thing, conversations say another. That’s not “multi-channel”—that’s a conflict to reconcile.\n\n**The best pre-meeting question:** “What changed in the *measurement* last week?” If nobody owns that answer, you’re about to debate fiction with confidence.",{"id":143,"name":144,"type":131,"typeVersion":16,"position":145,"parameters":146,"category":51,"deletable":8,"connectable":8},"node_text_auto","Guidance: automation vs judgment",[133,45],{"text":147},"**Trust automation when the rule is stable. Use human judgment when the world is not.**\n\n**Automation is great for:**\n- Repeating decisions with clear thresholds (eligibility, routing, alerts)\n- High-volume triage where being consistently “pretty good” beats being inconsistently brilliant\n- Enforcing definitions (same counting rules every time)\n\n**Humans still need to step in when:**\n- **Incentives are involved** (people will game what you measure)\n- The cost of a wrong decision is high (compliance, safety, large spend)\n- You’re in a **regime shift** (new branch manager, new offer, new channel, new system)\n\n**Practical guardrail:** keep a small weekly sample for human review. Automation should earn trust continuously, not once during launch week.",{"id":149,"name":150,"type":131,"typeVersion":16,"position":151,"parameters":152,"category":51,"deletable":8,"connectable":8},"node_text_messy","Guidance: messy evidence",[133,109],{"text":153},"**Don’t ‘clean’ your way into a lie.** The goal is decision-grade evidence, not cosmetically perfect dashboards.\n\n**How to keep the truth while reducing chaos:**\n- Separate **signal** (what happened) from **story** (why it happened). Don’t bake assumptions into fields.\n- Preserve raw notes/conversation context; summarize on top of it. When you delete nuance, you delete the explanation.\n- Track **unknowns as a category**. Forcing every record into a neat bucket creates fake certainty.\n- Use **directional confidence**: “We’re likely up, but attribution is noisy.” That’s a valid output.\n\n**Field tip:** if cleaning removes the uncomfortable cases first (timeouts, partial records, ambiguous attribution), you’re not cleaning—you’re curating.",{"id":155,"name":156,"type":131,"typeVersion":16,"position":157,"parameters":158,"category":51,"deletable":8,"connectable":8},"node_text_compare","Guidance: branch comparison",[133,115],{"text":159},"**Branch comparisons fail in the same predictable ways:**\n\n1) **Different denominators:** One branch has more walk-ins, another has more scheduled leads. Same ‘conversion’ label, different reality.\n2) **Different logging behavior:** The best-run branch often logs more issues (because they notice them). It can look worse on paper.\n3) **Attribution double-counting:** Conversations, form fills, and calls can all claim the same outcome. If totals exceed reality, you’re counting echoes.\n4) **Local policy differences:** hours, staffing, and eligibility rules make ‘performance’ non-comparable unless you normalize.\n\n**Make comparisons safer:**\n- Compare **like-for-like cohorts** (same lead type, same time window, same channel)\n- Pair outcome metrics with **process metrics** (speed-to-lead, follow-up attempts)\n- Always ask: “What would I conclude if attribution were 20% wrong?” If your decision flips, you need a sturdier signal.",{"id":161,"name":162,"type":131,"typeVersion":16,"position":163,"parameters":164,"category":51,"deletable":8,"connectable":8},"node_text_culture","Guidance: signal culture",[133,121],{"text":165},"**A good signal culture doesn’t produce more slides. It produces faster, calmer decisions.**\n\n**What works in practice:**\n- Establish a short list of **‘decision metrics’** (used to act) vs **‘narration metrics’** (used to explain). Confusing the two causes drama.\n- Assign **metric ownership**: every key number needs a person who can answer “what changed in measurement?”\n- Normalize saying **“unknown”** without penalty. If people fear looking uncertain, they’ll manufacture certainty.\n- Reward teams for surfacing tracking problems early—before leadership celebrates the wrong win.\n\n**The small witty rule:** if the meeting’s confidence rises faster than the data quality, you’re not aligning—you’re drifting.",{"id":167,"name":168,"type":55,"typeVersion":16,"position":169,"parameters":172,"category":51,"deletable":8,"connectable":8},"node_next_actions","Next actions",[170,171],1220,220,{"messageType":59,"headerText":168,"bodyText":173,"footerText":62,"sectionTitle":62,"buttons":174,"ctaDisplayText":62,"ctaUrl":62},"What would you like to do next?",[175,178],{"id":176,"title":177},"btn_next_more","More checkpoints",{"id":179,"title":84},"btn_next_handoff",{"id":181,"name":182,"type":88,"typeVersion":16,"position":183,"parameters":186,"category":94,"deletable":8,"connectable":8},"node_if_next_more","If: next = more checkpoints",[184,185],1460,200,{"buttonId":176,"operator":93},{"id":188,"name":189,"type":88,"typeVersion":16,"position":190,"parameters":192,"category":94,"deletable":8,"connectable":8},"node_if_next_handoff","If: next = handoff",[184,191],260,{"buttonId":179,"operator":93},{"id":194,"name":177,"type":55,"typeVersion":16,"position":195,"parameters":197,"category":51,"deletable":8,"connectable":8},"node_menu_2",[196,185],1700,{"messageType":59,"headerText":177,"bodyText":198,"footerText":62,"sectionTitle":62,"buttons":199,"ctaDisplayText":62,"ctaUrl":62},"Pick one more checkpoint. After this, we’ll wrap up (or I can hand you to an analyst).",[200,202,204,206,208,210,212],{"id":201,"title":66},"btn2_trust_numbers",{"id":203,"title":69},"btn2_dirty_signal",{"id":205,"title":72},"btn2_auto_vs_human",{"id":207,"title":75},"btn2_messy_evidence",{"id":209,"title":78},"btn2_compare_misreads",{"id":211,"title":81},"btn2_signal_culture",{"id":213,"title":84},"btn2_talk_to_analyst",{"id":215,"name":216,"type":88,"typeVersion":16,"position":217,"parameters":219,"category":94,"deletable":8,"connectable":8},"node_if2_trust","If2: trust numbers",[218,91],1940,{"buttonId":201,"operator":93},{"id":221,"name":222,"type":88,"typeVersion":16,"position":223,"parameters":224,"category":94,"deletable":8,"connectable":8},"node_if2_dirty","If2: dirty signal",[218,24],{"buttonId":203,"operator":93},{"id":226,"name":227,"type":88,"typeVersion":16,"position":228,"parameters":229,"category":94,"deletable":8,"connectable":8},"node_if2_auto","If2: automation vs judgment",[218,45],{"buttonId":205,"operator":93},{"id":231,"name":232,"type":88,"typeVersion":16,"position":233,"parameters":234,"category":94,"deletable":8,"connectable":8},"node_if2_messy","If2: messy evidence",[218,109],{"buttonId":207,"operator":93},{"id":236,"name":237,"type":88,"typeVersion":16,"position":238,"parameters":239,"category":94,"deletable":8,"connectable":8},"node_if2_compare","If2: branch comparison",[218,115],{"buttonId":209,"operator":93},{"id":241,"name":242,"type":88,"typeVersion":16,"position":243,"parameters":244,"category":94,"deletable":8,"connectable":8},"node_if2_culture","If2: signal culture",[218,121],{"buttonId":211,"operator":93},{"id":246,"name":247,"type":88,"typeVersion":16,"position":248,"parameters":249,"category":94,"deletable":8,"connectable":8},"node_if2_talk","If2: talk to analyst",[218,57],{"buttonId":213,"operator":93},{"id":251,"name":252,"type":131,"typeVersion":16,"position":253,"parameters":255,"category":51,"deletable":8,"connectable":8},"node_text2_trust","Guidance2: trust numbers",[254,91],2220,{"text":135},{"id":257,"name":258,"type":131,"typeVersion":16,"position":259,"parameters":260,"category":51,"deletable":8,"connectable":8},"node_text2_dirty","Guidance2: dirty signal",[254,24],{"text":141},{"id":262,"name":263,"type":131,"typeVersion":16,"position":264,"parameters":265,"category":51,"deletable":8,"connectable":8},"node_text2_auto","Guidance2: automation vs judgment",[254,45],{"text":147},{"id":267,"name":268,"type":131,"typeVersion":16,"position":269,"parameters":270,"category":51,"deletable":8,"connectable":8},"node_text2_messy","Guidance2: messy evidence",[254,109],{"text":153},{"id":272,"name":273,"type":131,"typeVersion":16,"position":274,"parameters":275,"category":51,"deletable":8,"connectable":8},"node_text2_compare","Guidance2: branch comparison",[254,115],{"text":159},{"id":277,"name":278,"type":131,"typeVersion":16,"position":279,"parameters":280,"category":51,"deletable":8,"connectable":8},"node_text2_culture","Guidance2: signal culture",[254,121],{"text":165},{"id":282,"name":283,"type":55,"typeVersion":16,"position":284,"parameters":286,"category":51,"deletable":8,"connectable":8},"node_finish_actions","Finish up",[285,171],2500,{"messageType":59,"headerText":283,"bodyText":287,"footerText":62,"sectionTitle":62,"buttons":288,"ctaDisplayText":62,"ctaUrl":62},"Want to wrap up, or hand this to an analyst?",[289,292],{"id":290,"title":291},"btn_finish_done","Done",{"id":293,"title":84},"btn_finish_handoff",{"id":295,"name":296,"type":88,"typeVersion":16,"position":297,"parameters":299,"category":94,"deletable":8,"connectable":8},"node_if_finish_done","If: finish = done",[298,185],2740,{"buttonId":290,"operator":93},{"id":301,"name":302,"type":88,"typeVersion":16,"position":303,"parameters":304,"category":94,"deletable":8,"connectable":8},"node_if_finish_handoff","If: finish = handoff",[298,191],{"buttonId":293,"operator":93},{"id":306,"name":307,"type":131,"typeVersion":16,"position":308,"parameters":310,"category":51,"deletable":8,"connectable":8},"node_text_done","Closeout",[309,185],2980,{"text":311},"All set. If you need another checkpoint later, start the flow again and pick the next decision you’re facing. (Your future self in the meeting will thank you.)",{"id":313,"name":314,"type":131,"typeVersion":16,"position":315,"parameters":317,"category":51,"deletable":8,"connectable":8},"node_text_unknown","Unknown selection",[133,316],480,{"text":318},"I didn’t catch a button selection. Please restart and choose one of the checkpoints so I can route you to the right truth test.",{"id":320,"name":321,"type":131,"typeVersion":16,"position":322,"parameters":323,"category":51,"deletable":8,"connectable":8},"node_text_next_unknown","Unknown next action",[196,121],{"text":324},"Please use the buttons so I can route you correctly: **More checkpoints** or **Talk to analyst**.",{"id":326,"name":327,"type":131,"typeVersion":16,"position":328,"parameters":329,"category":51,"deletable":8,"connectable":8},"node_text_unknown_2","Unknown selection (menu 2)",[254,316],{"text":330},"I didn’t catch a button selection. If you still need help, restart the flow and choose a checkpoint.",{"id":332,"name":333,"type":131,"typeVersion":16,"position":334,"parameters":336,"category":51,"deletable":8,"connectable":8},"node_text_finish_unknown","Unknown finish action",[309,335],320,{"text":337},"Please use the buttons: **Done** or **Talk to analyst**.",{"id":339,"name":340,"type":341,"typeVersion":16,"position":342,"parameters":343,"category":347,"deletable":8,"connectable":8},"node_fallback_analyst","Handoff: Data & Insights","fallback",[309,24],{"handoffMessage":344,"departmentId":345,"departmentName":346},"I’m handing you off to our Data & Insights team. To speed this up, please share: (1) the branch(es), (2) the metric(s) in question, (3) the time window, and (4) what decision you’re trying to make.","data-insights","Data & Insights","terminal",[349,354,356,358,361,364,366,368,370,372,374,376,378,380,382,384,386,388,390,392,394,396,398,400,402,404,406,408,410,412,414,416,418,420,422,424,426,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,458,460],{"id":350,"source":35,"target":41,"sourceHandle":351,"targetHandle":352,"type":353},"conn_input_kb","out","in","edge",{"id":355,"source":41,"target":53,"sourceHandle":351,"targetHandle":352,"type":353},"conn_kb_menu",{"id":357,"source":53,"target":86,"sourceHandle":351,"targetHandle":352,"type":353},"conn_menu_if_trust",{"id":359,"source":86,"target":129,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_trust_true","true",{"id":362,"source":86,"target":96,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_trust_false","false",{"id":365,"source":96,"target":137,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_dirty_true",{"id":367,"source":96,"target":101,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_dirty_false",{"id":369,"source":101,"target":143,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_auto_true",{"id":371,"source":101,"target":106,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_auto_false",{"id":373,"source":106,"target":149,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_messy_true",{"id":375,"source":106,"target":112,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_messy_false",{"id":377,"source":112,"target":155,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_compare_true",{"id":379,"source":112,"target":118,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_compare_false",{"id":381,"source":118,"target":161,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_culture_true",{"id":383,"source":118,"target":124,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_culture_false",{"id":385,"source":124,"target":339,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_talk_true",{"id":387,"source":124,"target":313,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_talk_false",{"id":389,"source":129,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_trust_to_next",{"id":391,"source":137,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_dirty_to_next",{"id":393,"source":143,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_auto_to_next",{"id":395,"source":149,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_messy_to_next",{"id":397,"source":155,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_compare_to_next",{"id":399,"source":161,"target":167,"sourceHandle":351,"targetHandle":352,"type":353},"conn_culture_to_next",{"id":401,"source":167,"target":181,"sourceHandle":351,"targetHandle":352,"type":353},"conn_next_to_if_more",{"id":403,"source":181,"target":194,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_next_more_true",{"id":405,"source":181,"target":188,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_next_more_false",{"id":407,"source":188,"target":339,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_next_handoff_true",{"id":409,"source":188,"target":320,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_next_handoff_false",{"id":411,"source":194,"target":215,"sourceHandle":351,"targetHandle":352,"type":353},"conn_menu2_if2_trust",{"id":413,"source":215,"target":251,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_trust_true",{"id":415,"source":215,"target":221,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_trust_false",{"id":417,"source":221,"target":257,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_dirty_true",{"id":419,"source":221,"target":226,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_dirty_false",{"id":421,"source":226,"target":262,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_auto_true",{"id":423,"source":226,"target":231,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_auto_false",{"id":425,"source":231,"target":267,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_messy_true",{"id":427,"source":231,"target":236,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_messy_false",{"id":429,"source":236,"target":272,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_compare_true",{"id":431,"source":236,"target":241,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_compare_false",{"id":433,"source":241,"target":277,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_culture_true",{"id":435,"source":241,"target":246,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_culture_false",{"id":437,"source":246,"target":339,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if2_talk_true",{"id":439,"source":246,"target":326,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if2_talk_false",{"id":441,"source":251,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_trust_to_finish",{"id":443,"source":257,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_dirty_to_finish",{"id":445,"source":262,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_auto_to_finish",{"id":447,"source":267,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_messy_to_finish",{"id":449,"source":272,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_compare_to_finish",{"id":451,"source":277,"target":282,"sourceHandle":351,"targetHandle":352,"type":353},"conn_text2_culture_to_finish",{"id":453,"source":282,"target":295,"sourceHandle":351,"targetHandle":352,"type":353},"conn_finish_to_if_done",{"id":455,"source":295,"target":306,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_finish_done_true",{"id":457,"source":295,"target":301,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_finish_done_false",{"id":459,"source":301,"target":339,"sourceHandle":360,"targetHandle":352,"type":353},"conn_if_finish_handoff_true",{"id":461,"source":301,"target":332,"sourceHandle":363,"targetHandle":352,"type":353},"conn_if_finish_handoff_false","automation",[28,30,29,464,465,466],"data-trust","attribution","ops-insights",[468,469],"WhatsApp","Web chat","intermediate","Calypso","2026-04-26T11:05:18.069Z","/en/workflows/signal-reliability-playbook-for-branch-decisions",{"en":473},{"title":9,"description":476,"ogDescription":477,"twitterDescription":478,"canonicalPath":473,"robots":479,"schemaType":480,"alternates":481},"Guide teams to trust the right branch signals, catch dirty data early, and decide when automation needs human review—via a simple chat flow.","A practical chat based playbook to test branch signal reliability, spot dirty data before meetings, and choose when automation needs human judgment.","Turn messy branch signals into decision ready insight: trust tests, dirty signal checks, and clear guidance on when automation needs human review.","index,follow","HowTo",[482],{"hreflang":6,"href":473},1778614431328]