{ "meta": { "templateCredsSetupCompleted": true }, "nodes": [ { "id": "193106dc-e2ae-4981-ba25-a2b7741f21be", "name": "When clicking 'Execute workflow'", "type": "n8n-nodes-base.manualTrigger", "position": [ 96, 496 ], "parameters": {}, "typeVersion": 1 }, { "id": "4782dc05-0b1c-4a96-bbb0-2563156981c5", "name": "Zoho CRM - Deal Trigger", "type": "n8n-nodes-base.zohoCrm", "position": [ 320, 496 ], "parameters": { "options": {}, "resource": "deal", "operation": "getAll" }, "typeVersion": 1 }, { "id": "846c23e1-ebf1-4e9b-85ab-a873d02201f1", "name": "Filter Recent Deals", "type": "n8n-nodes-base.code", "position": [ 544, 496 ], "parameters": { "jsCode": "// Filter: Only process deals closed in last 7 days to avoid spam\nconst sevenDaysAgo = new Date();\nsevenDaysAgo.setDate(sevenDaysAgo.getDate() - 7);\n\nconst filtered = items.filter(item => {\n const deal = item.json;\n const closedTime = deal.Closed_Time ? new Date(deal.Closed_Time) : null;\n const modifiedTime = deal.Modified_Time ? new Date(deal.Modified_Time) : null;\n \n // Use Closed_Time if available, otherwise Modified_Time\n const relevantTime = closedTime || modifiedTime;\n \n if (!relevantTime) return false;\n \n // Only process if closed/modified in last 7 days\n return relevantTime >= sevenDaysAgo;\n});\n\n// Limit to first 10 deals to prevent overwhelming Slack\nreturn filtered.slice(0, 10).map(item => ({ json: item.json }));" }, "typeVersion": 2 }, { "id": "909c5e09-f39f-4b27-bed4-e4f45553fcf4", "name": "Fetch Historical Averages", "type": "n8n-nodes-base.googleSheets", "position": [ 768, 496 ], "parameters": { "sheetName": { "__rl": true, "mode": "list", "value": "" }, "documentId": { "__rl": true, "mode": "list", "value": "" } }, "typeVersion": 3 }, { "id": "0d17d548-a0ef-432b-a9d3-ccad42321265", "name": "Analyze Cycle", "type": "n8n-nodes-base.code", "position": [ 992, 496 ], "parameters": { "jsCode": "// Get deal from Zoho trigger (first item from filtered deals)\nconst deal = $item(0).$node[\"Filter Recent Deals\"].json || $item(0).$node[\"Zoho CRM - Deal Trigger\"].json || {};\nconst created = deal.Created_Time ? new Date(deal.Created_Time) : null;\nconst closed = deal.Closed_Time ? new Date(deal.Closed_Time) : (deal.Modified_Time ? new Date(deal.Modified_Time) : null);\nconst totalDays = created && closed ? Math.max(1, Math.ceil((closed - created) / (1000 * 60 * 60 * 24))) : null;\n\n// Historical rows from Google Sheet (expects headers: Created_Time, Closed_Time, Stage, or similar)\nconst sheetRows = items.map(i => i.json).filter(r => r && (r.Created_Time || r['Created Time'] || r.created_time));\n\n// Normalize column names (handle different formats)\nconst normalizeRow = (row) => {\n return {\n Created_Time: row.Created_Time || row['Created Time'] || row.created_time || row.CreatedTime,\n Closed_Time: row.Closed_Time || row['Closed Time'] || row.closed_time || row.ClosedTime,\n Stage: row.Stage || row.stage || row.STAGE || row.Status || row.status,\n Deal_Name: row.Deal_Name || row['Deal Name'] || row.deal_name || row.DealName,\n Stage_History: row.Stage_History || row['Stage History'] || row.stage_history || null\n };\n};\n\nconst rows = sheetRows.map(normalizeRow).filter(r => r.Created_Time && r.Closed_Time);\n\n// Calculate cycle times from historical data\nconst cycles = rows.map(r => {\n try {\n const c = new Date(r.Created_Time);\n const cl = new Date(r.Closed_Time);\n if (isNaN(c.getTime()) || isNaN(cl.getTime())) return null;\n return Math.max(1, Math.ceil((cl - c) / (1000 * 60 * 60 * 24)));\n } catch (e) {\n return null;\n }\n}).filter(c => c !== null);\n\nconst avgDays = cycles.length ? +(cycles.reduce((a, b) => a + b, 0) / cycles.length).toFixed(1) : null;\nconst sortedCycles = [...cycles].sort((a, b) => a - b);\nconst medianDays = cycles.length ? sortedCycles[Math.floor(sortedCycles.length / 2)] : null;\nconst p90Days = cycles.length ? sortedCycles[Math.floor(sortedCycles.length * 0.9)] : null;\n\n// Best close day-of-week from historical wins\nconst winRows = rows.filter(r => {\n const stage = (r.Stage || '').toLowerCase();\n return stage.includes('won') || stage.includes('closed won');\n});\n\nconst dowCount = winRows.reduce((acc, r) => {\n try {\n const d = new Date(r.Closed_Time);\n if (isNaN(d.getTime())) return acc;\n const key = d.toLocaleDateString('en-US', { weekday: 'short' });\n acc[key] = (acc[key] || 0) + 1;\n } catch (e) {\n // Skip invalid dates\n }\n return acc;\n}, {});\n\nconst bestCloseDay = Object.entries(dowCount).sort((a, b) => b[1] - a[1])[0]?.[0] || null;\n\n// Stage-dwell analysis (if Stage_History is available as JSON string or comma-separated)\nlet longestDwellStage = null;\nlet longestDwellDays = 0;\n\nif (deal.Stage_History || deal['Stage History']) {\n try {\n const history = typeof deal.Stage_History === 'string' ? JSON.parse(deal.Stage_History) : deal.Stage_History;\n if (Array.isArray(history) && history.length > 1) {\n let maxDwell = 0;\n let maxStage = null;\n \n for (let i = 1; i < history.length; i++) {\n const prev = new Date(history[i-1].timestamp || history[i-1].date);\n const curr = new Date(history[i].timestamp || history[i].date);\n if (!isNaN(prev.getTime()) && !isNaN(curr.getTime())) {\n const dwell = Math.ceil((curr - prev) / (1000 * 60 * 60 * 24));\n if (dwell > maxDwell) {\n maxDwell = dwell;\n maxStage = history[i-1].stage || history[i-1].Stage || 'Unknown';\n }\n }\n }\n \n longestDwellStage = maxStage;\n longestDwellDays = maxDwell;\n }\n } catch (e) {\n // Stage history not in expected format, skip\n }\n}\n\n// Calculate average stage dwell from historical data (if available)\nlet avgStageDwells = {};\nif (rows.some(r => r.Stage_History)) {\n const stageDwells = {};\n rows.forEach(r => {\n try {\n if (r.Stage_History) {\n const history = typeof r.Stage_History === 'string' ? JSON.parse(r.Stage_History) : r.Stage_History;\n if (Array.isArray(history)) {\n for (let i = 1; i < history.length; i++) {\n const stage = history[i-1].stage || history[i-1].Stage || 'Unknown';\n const prev = new Date(history[i-1].timestamp || history[i-1].date);\n const curr = new Date(history[i].timestamp || history[i].date);\n if (!isNaN(prev.getTime()) && !isNaN(curr.getTime())) {\n const dwell = Math.ceil((curr - prev) / (1000 * 60 * 60 * 24));\n if (!stageDwells[stage]) stageDwells[stage] = [];\n stageDwells[stage].push(dwell);\n }\n }\n }\n }\n } catch (e) {\n // Skip invalid history\n }\n });\n \n Object.keys(stageDwells).forEach(stage => {\n const dwells = stageDwells[stage];\n avgStageDwells[stage] = +(dwells.reduce((a, b) => a + b, 0) / dwells.length).toFixed(1);\n });\n}\n\n// Build insights\nconst insights = [];\n\nif (totalDays && avgDays) {\n if (totalDays > avgDays * 1.25) {\n insights.push(`Slow cycle: ${totalDays}d vs avg ${avgDays}d (+${(totalDays - avgDays).toFixed(1)}d)`);\n } else if (totalDays < avgDays * 0.75) {\n insights.push(`Fast cycle: ${totalDays}d vs avg ${avgDays}d (-${(avgDays - totalDays).toFixed(1)}d)`);\n } else {\n insights.push(`Normal cycle: ${totalDays}d (avg ${avgDays}d)`);\n }\n}\n\nif (medianDays) insights.push(`Median historical cycle: ${medianDays}d`);\nif (p90Days) insights.push(`90th percentile: ${p90Days}d`);\nif (bestCloseDay) insights.push(`Best close day: ${bestCloseDay}`);\n\nif (longestDwellStage && longestDwellDays > 0) {\n insights.push(`Longest stage dwell: ${longestDwellStage} (${longestDwellDays}d)`);\n \n // Compare to historical average for that stage\n if (avgStageDwells[longestDwellStage]) {\n const avgDwell = avgStageDwells[longestDwellStage];\n if (longestDwellDays > avgDwell * 1.5) {\n insights.push(`${longestDwellStage} took ${longestDwellDays}d (avg ${avgDwell}d) - bottleneck detected!`);\n }\n }\n}\n\n// Process improvement suggestions\nconst suggestions = [];\nif (totalDays && avgDays && totalDays > avgDays * 1.25) {\n suggestions.push('Schedule immediate follow-up within 24h to unblock progress');\n}\nif (longestDwellStage && longestDwellDays > 7) {\n suggestions.push(`Review ${longestDwellStage} stage process - consider early alignment with approval/legal`);\n}\nif (bestCloseDay) {\n suggestions.push(`Replicate patterns from ${bestCloseDay} closes for future deals`);\n}\nif (p90Days && totalDays && totalDays > p90Days) {\n suggestions.push('This deal exceeded 90th percentile - conduct post-mortem review');\n}\nif (suggestions.length === 0) {\n suggestions.push('Continue current process - performance is within expected range');\n}\n\nif (!insights.length) insights.push(' No historical insights available - check sheet data format');\n\nreturn [{ \n json: { \n deal, \n totalDays, \n avgDays, \n medianDays, \n p90Days, \n bestCloseDay, \n longestDwellStage,\n longestDwellDays,\n insights,\n suggestions,\n hasData: cycles.length > 0\n } \n}];" }, "typeVersion": 2 }, { "id": "8edc9a91-451f-4747-9cf4-9405928b9843", "name": "Check Valid Data", "type": "n8n-nodes-base.if", "position": [ 2112, 496 ], "parameters": { "options": {}, "conditions": { "options": { "leftValue": "", "caseSensitive": true, "typeValidation": "strict" }, "combinator": "and", "conditions": [ { "id": "check-deal-data", "operator": { "type": "string", "operation": "isNotEmpty" }, "leftValue": "={{ $json.deal && $json.deal.Deal_Name }}", "rightValue": "" }, { "id": "check-cycle-data", "operator": { "type": "number", "operation": "larger" }, "leftValue": "={{ $json.totalDays }}", "rightValue": 0 } ] } }, "typeVersion": 2.2 }, { "id": "25a13331-afb3-4897-8467-bfa066ff1c43", "name": "Slack Notification", "type": "n8n-nodes-base.slack", "position": [ 2336, 400 ], "parameters": { "text": "=*AI-Powered Sales Cycle Insights*\n\n*Deal:* {{$json.deal.Deal_Name || 'Unknown'}}\n*Owner:* {{$json.deal.Owner?.name || $json.deal.Owner || 'Unassigned'}}\n*Outcome:* {{$json.deal.Stage}}\n*Value:* {{$json.deal.Amount || 'N/A'}}\n\n*Cycle Analysis:*\n\u2022 Total Cycle: *{{$json.totalDays || 'n/a'}} days* (Avg: {{$json.avgDays || 'n/a'}}d)\n\u2022 Performance Score: *{{$json.performanceScore || 'N/A'}}/100*\n\u2022 Risk Level: *{{$json.riskLevel || 'Unknown'}}*\n{{ $json.insights ? $json.insights.map(i => '\u2022 ' + i).join('\\n') : '\u2022 No insights available' }}\n\n*AI Intelligence:*\n\u2022 Sentiment: *{{$json.sentimentAnalysis?.sentiment || $json.sentimentAnalysis || 'N/A'}}*\n\u2022 Win Probability: *{{$json.predictiveAnalytics?.winProbability || $json.predictiveAnalytics || 'N/A'}}*\n\u2022 Expected Close: *{{$json.predictiveAnalytics?.expectedCloseDate || 'N/A'}}*\n\n*AI Smart Recommendations:*\n{{$json.text || ($json.suggestions ? $json.suggestions.map(s => '\u2022 ' + s).join('\\n') : '\u2022 Processing recommendations...') }}\n\n*Visualizations:* {{$json.chartSummary || 'Charts generated'}}\n\n---\n_Generated: {{$now}} | AI-Enhanced Analysis_", "channel": "#sales-insights", "attachments": [], "otherOptions": {} }, "typeVersion": 1 }, { "id": "2168b744-8893-4922-8e92-b4717ff62fe2", "name": "Log to Historical Sheet", "type": "n8n-nodes-base.googleSheets", "position": [ 2336, 592 ], "parameters": { "operation": "append", "sheetName": { "__rl": true, "mode": "list", "value": "" }, "documentId": { "__rl": true, "mode": "list", "value": "" } }, "typeVersion": 3 }, { "id": "69ae8a3f-e21e-473c-8e47-e66bc4e0ccc0", "name": "Main Overview", "type": "n8n-nodes-base.stickyNote", "position": [ -320, -64 ], "parameters": { "color": "yellow", "width": 448, "height": 496, "content": "## How it works\n\nThis workflow automatically analyzes sales cycle performance for closed deals (won/lost) from Zoho CRM. It compares each deal to historical averages, identifies bottlenecks, calculates optimal timing patterns, and suggests process improvements. Results are sent to Slack and logged to Google Sheets for continuous optimization.\n\n## Setup steps\n\n1. Configure Zoho CRM credentials in the \"Zoho CRM - Deal Trigger\" node\n2. Set up Google Sheets credentials and configure \"Fetch Historical Averages\" with your sheet ID and range\n3. Configure \"Log to Historical Sheet\" to append new deals to your historical data\n4. Update Slack channel in \"Slack Notification\" node (default: #sales-insights)\n5. Ensure your Google Sheet has headers: Created_Time, Closed_Time, Stage, Deal_Name (or similar)\n6. Optionally replace Manual Trigger with Schedule Trigger for automatic daily/hourly runs" }, "typeVersion": 1 }, { "id": "fdc158ac-d7e8-49a0-a756-b1012e00e493", "name": "Section - Data Collection", "type": "n8n-nodes-base.stickyNote", "position": [ 256, 192 ], "parameters": { "color": 7, "width": 656, "height": 560, "content": "## Data Collection\n\nFetches closed deals (won/lost) from Zoho CRM, filters to recent deals (last 7 days, max 10), and retrieves historical cycle data from Google Sheets for comparison and analysis." }, "typeVersion": 1 }, { "id": "07e73147-cbed-4229-86be-cfc5cbd57aa1", "name": "AI Sentiment Analysis", "type": "n8n-nodes-base.openAi", "position": [ 1216, 400 ], "parameters": { "model": "gpt-4", "prompt": "Analyze the sentiment and emotional tone of this deal information. Provide: 1) Overall sentiment (positive/neutral/negative), 2) Confidence score (0-100%), 3) Key emotional indicators, 4) Risk level based on sentiment. Deal Data: {{ JSON.stringify($json.deal) }}", "options": {}, "requestOptions": {} }, "typeVersion": 1 }, { "id": "f3ae52ce-85c1-4dab-a121-1e6fd9806af9", "name": "AI Predictive Analytics", "type": "n8n-nodes-base.openAi", "position": [ 1216, 592 ], "parameters": { "model": "gpt-4", "prompt": "Based on historical deal patterns and current deal data, predict the outcome probability. Analyze: 1) Win probability (0-100%), 2) Expected close date, 3) Risk factors, 4) Confidence level. Historical Context: {{ JSON.stringify($json.historicalData) }} Current Deal: {{ JSON.stringify($json.deal) }}", "options": {}, "requestOptions": {} }, "typeVersion": 1 }, { "id": "fcd95efa-793c-46a8-a940-7e05f51d330d", "name": "AI Smart Recommendations", "type": "n8n-nodes-base.openAi", "position": [ 1888, 496 ], "parameters": { "model": "gpt-4", "prompt": "You are a sales intelligence expert. Based on the comprehensive analysis below, generate 3-5 prioritized, actionable recommendations. Consider: 1) Historical patterns, 2) Sentiment analysis insights, 3) Predictive analytics, 4) Performance metrics, 5) Risk factors. Format each recommendation with: [PRIORITY] Recommendation text (Expected Impact).\n\nDeal Analysis: {{ JSON.stringify($json) }}\n\nSentiment Analysis: {{ $json.sentimentAnalysis || 'Not available' }}\n\nPredictive Analytics: {{ $json.predictiveAnalytics || 'Not available' }}", "options": {}, "requestOptions": {} }, "typeVersion": 1 }, { "id": "156892fd-de90-4c3d-9278-6db6e01fb279", "name": "AI Data Aggregation", "type": "n8n-nodes-base.code", "position": [ 1440, 496 ], "parameters": { "jsCode": "// MERGE all AI outputs before sending to recommendations\n// This node aggregates data from parallel AI branches\nconst mainData = $input.first().json || {};\nconst sentimentData = $input.item(1)?.json || {};\nconst predictiveData = $input.item(2)?.json || {};\n\n// Extract AI responses\nlet sentimentAnalysis = sentimentData.text || sentimentData.sentimentAnalysis || sentimentData;\nlet predictiveAnalytics = predictiveData.text || predictiveData.predictiveAnalytics || predictiveData;\n\n// Try to parse JSON responses\ntry {\n if (typeof sentimentAnalysis === 'string') {\n const parsed = JSON.parse(sentimentAnalysis);\n sentimentAnalysis = parsed;\n }\n} catch (e) {\n // Keep as string\n}\n\ntry {\n if (typeof predictiveAnalytics === 'string') {\n const parsed = JSON.parse(predictiveAnalytics);\n predictiveAnalytics = parsed;\n }\n} catch (e) {\n // Keep as string\n}\n\n// Create merged dataset for AI Smart Recommendations\nconst mergedData = {\n ...mainData,\n sentimentAnalysis: sentimentAnalysis,\n predictiveAnalytics: predictiveAnalytics,\n // Ensure suggestions array exists for Slack template\n suggestions: mainData.suggestions || []\n};\n\nreturn { json: mergedData };" }, "typeVersion": 2 }, { "id": "d3c41db7-516e-4b4f-977f-5c485c5d5adc", "name": "Advanced Data Visualization", "type": "n8n-nodes-base.code", "position": [ 1664, 496 ], "parameters": { "jsCode": "// Generate advanced data visualization data for charts and trends\nconst deal = $json.deal || {};\nconst historical = $json.historicalData || [];\n\n// Prepare data for different chart types\nconst visualizationData = {\n // Cycle time trend over time\n cycleTimeTrend: historical.map((item, index) => ({\n period: `Period ${index + 1}`,\n cycleTime: item.totalDays || 0,\n dealValue: item.deal?.Amount || 0\n })),\n \n // Stage distribution\n stageDistribution: historical.reduce((acc, item) => {\n const stage = item.deal?.Stage || 'Unknown';\n acc[stage] = (acc[stage] || 0) + 1;\n return acc;\n }, {}),\n \n // Performance metrics\n performanceMetrics: {\n avgCycleTime: $json.avgDays || 0,\n currentCycleTime: $json.totalDays || 0,\n winRate: historical.filter(h => h.deal?.Stage?.toLowerCase().includes('won')).length / historical.length * 100,\n totalValue: historical.reduce((sum, h) => sum + (h.deal?.Amount || 0), 0)\n },\n \n // Risk heatmap data\n riskFactors: {\n sentimentRisk: $json.sentimentAnalysis?.riskLevel || 'Low',\n cycleRisk: $json.totalDays > ($json.avgDays * 1.5) ? 'High' : 'Low',\n predictionRisk: $json.predictiveAnalytics?.winProbability < 50 ? 'High' : 'Low'\n }\n};\n\n// Generate chart configurations\nconst charts = {\n cycleTrendChart: {\n type: 'line',\n data: visualizationData.cycleTimeTrend,\n title: 'Sales Cycle Time Trend'\n },\n stagePieChart: {\n type: 'pie',\n data: Object.entries(visualizationData.stageDistribution).map(([stage, count]) => ({\n label: stage,\n value: count\n })),\n title: 'Deal Distribution by Stage'\n },\n performanceGauge: {\n type: 'gauge',\n data: visualizationData.performanceMetrics.winRate,\n title: 'Win Rate %'\n }\n};\n\nreturn {\n json: {\n ...$json,\n visualizationData,\n charts,\n chartSummary: `Generated ${Object.keys(charts).length} visualizations: cycle trends, stage distribution, and performance metrics`\n }\n};" }, "typeVersion": 2 }, { "id": "6c70e97f-a929-4a29-b0ac-16d3b1d88b46", "name": "Section - AI Intelligence", "type": "n8n-nodes-base.stickyNote", "position": [ 960, 192 ], "parameters": { "color": 7, "width": 1056, "height": 560, "content": "## AI-Powered Intelligence Layer\n\nThis section adds advanced AI capabilities: sentiment analysis for emotional intelligence, predictive analytics for outcome forecasting, smart recommendations for process optimization, and data visualization for trend analysis." }, "typeVersion": 1 }, { "id": "8b85fed8-8944-4d43-a381-021738147b2f", "name": "Section - Notifications & Logging", "type": "n8n-nodes-base.stickyNote", "position": [ 2064, 192 ], "parameters": { "color": 7, "width": 544, "height": 560, "content": "## Notifications & Logging\n\nValidates data quality, then sends formatted insights to Slack channel and automatically logs the analyzed deal to Google Sheets for building historical dataset over time." }, "typeVersion": 1 } ], "connections": { "Analyze Cycle": { "main": [ [ { "node": "AI Sentiment Analysis", "type": "main", "index": 0 }, { "node": "AI Predictive Analytics", "type": "main", "index": 0 } ] ] }, "Check Valid Data": { "main": [ [ { "node": "Slack Notification", "type": "main", "index": 0 }, { "node": "Log to Historical Sheet", "type": "main", "index": 0 } ] ] }, "AI Data Aggregation": { "main": [ [ { "node": "Advanced Data Visualization", "type": "main", "index": 0 } ] ] }, "Filter Recent Deals": { "main": [ [ { "node": "Fetch Historical Averages", "type": "main", "index": 0 } ] ] }, "AI Sentiment Analysis": { "main": [ [ { "node": "AI Data Aggregation", "type": "main", "index": 0 } ] ] }, "AI Predictive Analytics": { "main": [ [ { "node": "AI Data Aggregation", "type": "main", "index": 0 } ] ] }, "Zoho CRM - Deal Trigger": { "main": [ [ { "node": "Filter Recent Deals", "type": "main", "index": 0 } ] ] }, "AI Smart Recommendations": { "main": [ [ { "node": "Check Valid Data", "type": "main", "index": 0 } ] ] }, "Fetch Historical Averages": { "main": [ [ { "node": "Analyze Cycle", "type": "main", "index": 0 } ] ] }, "Advanced Data Visualization": { "main": [ [ { "node": "AI Smart Recommendations", "type": "main", "index": 0 } ] ] }, "When clicking 'Execute workflow'": { "main": [ [ { "node": "Zoho CRM - Deal Trigger", "type": "main", "index": 0 } ] ] } } }