Análisis de centro de llamadas (doble validación de IA con modelos de DeepSeek)

Intermedio

Este es unCRM, AI Summarizationflujo de automatización del dominio deautomatización que contiene 15 nodos.Utiliza principalmente nodos como Code, Webhook, HttpRequest, ManualTrigger, ChainLlm. Análisis de centro de llamadas con doble verificación de IA utilizando el modelo DeepSeek

Requisitos previos
  • Punto final de HTTP Webhook (n8n generará automáticamente)
  • Pueden requerirse credenciales de autenticación para la API de destino
Vista previa del flujo de trabajo
Visualización de las conexiones entre nodos, con soporte para zoom y panorámica
Exportar flujo de trabajo
Copie la siguiente configuración JSON en n8n para importar y usar este flujo de trabajo
{
  "meta": {
    "instanceId": "66ce8bb89c7868f862e0d2e755cd17c6a5aea7904e5504a5b2e292e317980443",
    "templateCredsSetupCompleted": true
  },
  "nodes": [
    {
      "id": "613422d5-05db-4163-bcb0-3fdae9de260b",
      "name": "Al hacer clic en 'Probar flujo de trabajo'",
      "type": "n8n-nodes-base.manualTrigger",
      "position": [
        -1120,
        60
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "9ec28d5b-6ea0-4a57-911e-9f4546b739a2",
      "name": "Nota adhesiva",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -520,
        -240
      ],
      "parameters": {
        "color": 7,
        "width": 340,
        "height": 440,
        "content": "## Generate report\nUsing deepseek R1"
      },
      "typeVersion": 1
    },
    {
      "id": "cf30eb86-2aad-4d33-a5c2-737239e63636",
      "name": "Nota adhesiva1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -140,
        -240
      ],
      "parameters": {
        "color": 7,
        "width": 340,
        "height": 440,
        "content": "## Double-check\nUsing deepseek V3"
      },
      "typeVersion": 1
    },
    {
      "id": "d799760f-83c5-4603-8eb8-3857807b364a",
      "name": "HTTP Solicitud",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        300,
        -140
      ],
      "parameters": {
        "url": "YOUR_CALL_BACK_API",
        "method": "POST",
        "options": {},
        "jsonBody": "={\n  data: \"{{$node['Report'].json.text}}\"\n}",
        "sendBody": true,
        "specifyBody": "json"
      },
      "typeVersion": 4.2
    },
    {
      "id": "bf0a8ab6-06e0-4564-96af-ddcf6795a845",
      "name": "Informe",
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "position": [
        -480,
        -140
      ],
      "parameters": {
        "text": "=You are a CRM data analyst assistant. Your task is to analyze the provided CRM data and generate valuable insights in Markdown format.\n\nYou will receive JSON data extracted from a CRM system that may include information about:\n- Call canter agents metrics.\n\n# ANALYSIS REQUIREMENTS\nAnalyze the data considering:\n1. Lead conversion rates and quality metrics\n2. Upsall\n3. Rank the agents with small description about every one.\n\n# OUTPUT FORMAT\nStructure your analysis in Markdown.\n\n# GUIDELINES\n- Focus on actionable insights rather than just describing the data\n- Use bullet points and tables when appropriate to improve readability\n- Include both positive findings and areas for improvement\n- Reference specific data points to support your analysis\n- Prioritize quality over quantity in your recommendations\n- Be concise yet thorough\n- If there are data quality issues or missing information, note these limitations\n- If you detect any unusual patterns or anomalies, highlight them\n\n# DATA\n```\n{{ JSON.stringify($input.first().json.body) }}\n```",
        "promptType": "define"
      },
      "typeVersion": 1.6
    },
    {
      "id": "f51d601e-898c-4dd6-894b-2e4911a334db",
      "name": "DeepSeek Razonamiento",
      "type": "@n8n/n8n-nodes-langchain.lmChatDeepSeek",
      "position": [
        -400,
        60
      ],
      "parameters": {
        "model": "deepseek-reasoner",
        "options": {}
      },
      "credentials": {
        "deepSeekApi": {
          "id": "ltpFxb7M3kHaEBFD",
          "name": "DeepSeek account"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "9393be48-aebc-4f6c-b445-014633e0e289",
      "name": "DeepSeek Chat",
      "type": "@n8n/n8n-nodes-langchain.lmChatDeepSeek",
      "position": [
        -20,
        80
      ],
      "parameters": {
        "options": {}
      },
      "credentials": {
        "deepSeekApi": {
          "id": "ltpFxb7M3kHaEBFD",
          "name": "DeepSeek account"
        }
      },
      "typeVersion": 1
    },
    {
      "id": "4a8dc360-fd3f-46a5-89e1-87ece59b0bb6",
      "name": "Datos de ejemplo",
      "type": "n8n-nodes-base.code",
      "position": [
        -860,
        60
      ],
      "parameters": {
        "jsCode": "return {\n  \"body\": \n    // You can use any data as JSON\n    // this is just example\n    // data start here\n    [\n      {\n        \"user\": {\n          \"id\": 15,\n          \"full_name\": \"lisa confirmation\",\n        },\n        \"productivity\": 44.67,\n        \"total_leads\": 465,\n        \"total_confirmed\": 291,\n        \"total_delivred\": 130,\n        \"total_in_proccess\": 119,\n        \"total_cancled\": 0,\n        \"total_returned\": 13,\n        \"total_assign\": 495,\n        \"total_need_confirmation\": 0,\n        \"total_recheck\": 22,\n        \"upsell\": 59,\n        \"upsell_delivered\": 27,\n        \"confirmation_rate\": 62.58\n      },\n      {\n        \"user\": {\n          \"id\": 1346,\n          \"full_name\": \"Sallam Confirmation\",\n        },\n        \"productivity\": 42.29,\n        \"total_leads\": 374,\n        \"total_confirmed\": 253,\n        \"total_delivred\": 107,\n        \"total_in_proccess\": 96,\n        \"total_cancled\": 0,\n        \"total_returned\": 21,\n        \"total_assign\": 459,\n        \"total_need_confirmation\": 1,\n        \"total_recheck\": 1,\n        \"upsell\": 62,\n        \"upsell_delivered\": 31,\n        \"confirmation_rate\": 67.65\n      }\n    ]\n    // data end here\n}"
      },
      "typeVersion": 2
    },
    {
      "id": "41e36370-fa7a-4f6e-a439-15127dfc432d",
      "name": "Nota adhesiva2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1160,
        -20
      ],
      "parameters": {
        "width": 480,
        "height": 240,
        "content": "## Test Workflow\nClick this button to test the workflow with example data"
      },
      "typeVersion": 1
    },
    {
      "id": "e0d473e8-7a2e-4473-9af7-5f2f835990db",
      "name": "Nota adhesiva8",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1160,
        240
      ],
      "parameters": {
        "width": 480,
        "height": 80,
        "content": "## Just to test"
      },
      "typeVersion": 1
    },
    {
      "id": "b0c31c3b-1fd0-485e-a54b-9a3045bcf09e",
      "name": "Nota adhesiva7",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1160,
        400
      ],
      "parameters": {
        "color": 4,
        "width": 1540,
        "height": 100,
        "content": "## Do you need more help or have any suggestions?\nContact me at mediaplus.ma@gmail.com"
      },
      "typeVersion": 1
    },
    {
      "id": "9a1f03ee-167f-44e3-ae12-61ab8d3789f2",
      "name": "Nota adhesiva6",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -520,
        -360
      ],
      "parameters": {
        "color": 7,
        "width": 720,
        "height": 80,
        "content": "## Change here\nYou can edit/add details about your goal by changing the AI promps.\n"
      },
      "typeVersion": 1
    },
    {
      "id": "d91de9c8-835b-4232-a76b-e27554ad595d",
      "name": "Disparador Webhook",
      "type": "n8n-nodes-base.webhook",
      "position": [
        -980,
        -300
      ],
      "webhookId": "b408defb-315d-4676-b4c4-1dcebe81ffc0",
      "parameters": {
        "path": "b408defb-315d-4676-b4c4-1dcebe81ffc0",
        "options": {},
        "httpMethod": [
          "POST",
          "GET"
        ],
        "multipleMethods": true
      },
      "typeVersion": 2
    },
    {
      "id": "04c2da18-10c9-44df-8084-52b5901ecc18",
      "name": "Nota adhesiva9",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1020,
        -380
      ],
      "parameters": {
        "color": 4,
        "width": 200,
        "height": 240,
        "content": "## Production"
      },
      "typeVersion": 1
    },
    {
      "id": "a1fb1cbc-391c-4918-b48f-8b44116921b8",
      "name": "Revisar",
      "type": "@n8n/n8n-nodes-langchain.chainLlm",
      "position": [
        -100,
        -140
      ],
      "parameters": {
        "text": "=You are a Data Analysis Verification Expert. Your task is to evaluate whether an AI-generated report accurately and completely analyzes the provided CRM data. You will assess the report quality and determine if it's compatible with the original input.\n\n# INPUT\nYou will receive:\n1. The original call center agent metrics data (JSON)\n2. The AI-generated analysis report in Markdown\n\n# VERIFICATION REQUIREMENTS\nEvaluate the report for:\n1. Factual accuracy - Do all numbers, rankings, and statements accurately reflect the data?\n2. Comprehensiveness - Does the report cover all required areas? (Lead conversion, Upsell, Agent ranking)\n3. Insight quality - Does the report provide meaningful insights beyond basic data description?\n4. Completeness - Are all agents included in the analysis?\n5. Format compliance - Is the report properly formatted in Markdown with appropriate sections?\n\n# OUTPUT FORMAT\nReturn a JSON object with the following structure:\n```json\n{\n  \"verified\": true/false,\n  \"score\": 1-10,\n  \"quality_assessment\": \"Brief 2-4 sentence evaluation of report quality\",\n  \"missing_elements\": [\"List any required elements missing from the report\"],\n  \"inaccuracies\": [\"List any factual errors or misinterpretations\"],\n  \"improvement_suggestions\": [\"Specific suggestions for report improvement\"]\n}\n```\n\n# EVALUATION CRITERIA\n- \"verified\": Set to true ONLY if the report is factually accurate, includes all agents, covers all required areas, and provides meaningful insights.\n- \"score\": Rate from 1-10 where:\n  * 1-3: Poor report with major inaccuracies or missing elements\n  * 4-6: Adequate report with some issues\n  * 7-8: Good report with minor issues\n  * 9-10: Excellent report with comprehensive analysis\n\n# GUIDELINES\n- Be thorough and precise in your verification\n- Check all numerical claims against the original data\n- Verify that all agents are properly ranked and described\n- Check that lead conversion rates and upsell metrics are accurately analyzed\n- Assess whether the insights are actionable and valuable\n- Maintain a balanced perspective, noting both strengths and weaknesses\n\n# ORIGINAL DATA\n{{ JSON.stringify($node[\"Example data\"].json.chatInput) }}\n\n# AI-GENERATED REPORT\n{{ $json.text }}",
        "promptType": "define"
      },
      "typeVersion": 1.6
    }
  ],
  "pinData": {},
  "connections": {
    "bf0a8ab6-06e0-4564-96af-ddcf6795a845": {
      "main": [
        [
          {
            "node": "a1fb1cbc-391c-4918-b48f-8b44116921b8",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "a1fb1cbc-391c-4918-b48f-8b44116921b8": {
      "main": [
        [
          {
            "node": "d799760f-83c5-4603-8eb8-3857807b364a",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Webhook": {
      "main": [
        [
          {
            "node": "bf0a8ab6-06e0-4564-96af-ddcf6795a845",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "4a8dc360-fd3f-46a5-89e1-87ece59b0bb6": {
      "main": [
        [
          {
            "node": "bf0a8ab6-06e0-4564-96af-ddcf6795a845",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "9393be48-aebc-4f6c-b445-014633e0e289": {
      "ai_languageModel": [
        [
          {
            "node": "a1fb1cbc-391c-4918-b48f-8b44116921b8",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "f51d601e-898c-4dd6-894b-2e4911a334db": {
      "ai_languageModel": [
        [
          {
            "node": "bf0a8ab6-06e0-4564-96af-ddcf6795a845",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "613422d5-05db-4163-bcb0-3fdae9de260b": {
      "main": [
        [
          {
            "node": "4a8dc360-fd3f-46a5-89e1-87ece59b0bb6",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}
Preguntas frecuentes

¿Cómo usar este flujo de trabajo?

Copie el código de configuración JSON de arriba, cree un nuevo flujo de trabajo en su instancia de n8n y seleccione "Importar desde JSON", pegue la configuración y luego modifique la configuración de credenciales según sea necesario.

¿En qué escenarios es adecuado este flujo de trabajo?

Intermedio - CRM, Resumen de IA

¿Es de pago?

Este flujo de trabajo es completamente gratuito, puede importarlo y usarlo directamente. Sin embargo, tenga en cuenta que los servicios de terceros utilizados en el flujo de trabajo (como la API de OpenAI) pueden requerir un pago por su cuenta.

Información del flujo de trabajo
Nivel de dificultad
Intermedio
Número de nodos15
Categoría2
Tipos de nodos7
Descripción de la dificultad

Adecuado para usuarios con experiencia intermedia, flujos de trabajo de complejidad media con 6-15 nodos

Autor
Omar Akoudad

Omar Akoudad

@mediaplusma

Automation, Code, and Analytics for E-commerce businesses, We help businesses streamline operations using n8n, AI agents, and data science to enhance efficiency and scalability.

Enlaces externos
Ver en n8n.io

Compartir este flujo de trabajo

Categorías

Categorías: 34