Automatisation WhatsApp avec n8n : agent de vente AI en temps réel
- Ce workflow n8n est conçu pour automatiser les interactions sur WhatsApp en intégrant un agent de vente alimenté par l'intelligence artificielle. Dans un contexte commercial où la réactivité et la personnalisation des échanges sont essentielles, ce workflow permet de répondre efficacement aux demandes des clients tout en optimisant le temps de traitement des informations. En utilisant des outils comme OpenAI et des modèles d'embeddings, il offre une solution innovante pour les entreprises souhaitant améliorer leur service client via WhatsApp.
- Le processus débute avec le déclencheur 'WhatsApp Trigger', qui capte les messages entrants. Ensuite, les données sont traitées par plusieurs nœuds, notamment 'OpenAI Chat Model' et 'Embeddings OpenAI', qui analysent et génèrent des réponses adaptées. Le 'Window Buffer Memory' permet de garder en mémoire le contexte des conversations, tandis que le 'Vector Store Tool' et le 'Product Catalogue' facilitent l'accès aux informations produits. La gestion des types de messages est assurée par le nœud 'Handle Message Types', qui dirige les requêtes vers les réponses appropriées. Enfin, les réponses sont envoyées via le nœud 'Reply To User', garantissant une interaction fluide et naturelle.
- Les bénéfices de ce workflow sont multiples : il réduit le temps de réponse, améliore la satisfaction client et permet aux équipes de vente de se concentrer sur des tâches à plus forte valeur ajoutée. En intégrant ce type d'automatisation n8n, les entreprises peuvent non seulement augmenter leur efficacité opérationnelle, mais aussi renforcer leur relation client.
Workflow n8n WhatsApp, intelligence artificielle, service client : vue d'ensemble
Schéma des nœuds et connexions de ce workflow n8n, généré à partir du JSON n8n.
Workflow n8n WhatsApp, intelligence artificielle, service client : détail des nœuds
Inscris-toi pour voir l'intégralité du workflow
Inscription gratuite
S'inscrire gratuitementBesoin d'aide ?{
"meta": {
"instanceId": "408f9fb9940c3cb18ffdef0e0150fe342d6e655c3a9fac21f0f644e8bedabcd9"
},
"nodes": [
{
"id": "77ee6494-4898-47dc-81d9-35daf6f0beea",
"name": "WhatsApp Trigger",
"type": "n8n-nodes-base.whatsAppTrigger",
"position": [
1360,
-280
],
"webhookId": "aaa71f03-f7af-4d18-8d9a-0afb86f1b554",
"parameters": {
"updates": [
"messages"
]
},
"credentials": {
"whatsAppTriggerApi": {
"id": "H3uYNtpeczKMqtYm",
"name": "WhatsApp OAuth account"
}
},
"typeVersion": 1
},
{
"id": "57210e27-1f89-465a-98cc-43f890a4bf58",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1960,
-200
],
"parameters": {
"model": "gpt-4o-2024-08-06",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "e1053235-0ade-4e36-9ad2-8b29c78fced8",
"name": "Window Buffer Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
2080,
-200
],
"parameters": {
"sessionKey": "=whatsapp-75-{{ $json.messages[0].from }}",
"sessionIdType": "customKey"
},
"typeVersion": 1.2
},
{
"id": "69f1b78b-7c93-4713-863a-27e04809996f",
"name": "Vector Store Tool",
"type": "@n8n/n8n-nodes-langchain.toolVectorStore",
"position": [
2200,
-200
],
"parameters": {
"name": "query_product_brochure",
"description": "Call this tool to query the product brochure. Valid for the year 2024."
},
"typeVersion": 1
},
{
"id": "170e8f7d-7e14-48dd-9f80-5352cc411fc1",
"name": "Embeddings OpenAI",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"position": [
2200,
80
],
"parameters": {
"model": "text-embedding-3-small",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "ee78320b-d407-49e8-b4b8-417582a44709",
"name": "OpenAI Chat Model1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
2440,
-60
],
"parameters": {
"model": "gpt-4o-2024-08-06",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "9dd89378-5acf-4ca6-8d84-e6e64254ed02",
"name": "When clicking ‘Test workflow’",
"type": "n8n-nodes-base.manualTrigger",
"position": [
0,
-240
],
"parameters": {},
"typeVersion": 1
},
{
"id": "e68fc137-1bcb-43f0-b597-3ae07f380c15",
"name": "Embeddings OpenAI1",
"type": "@n8n/n8n-nodes-langchain.embeddingsOpenAi",
"position": [
760,
-20
],
"parameters": {
"model": "text-embedding-3-small",
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "2d31e92b-18d4-4f6b-8cdb-bed0056d50d7",
"name": "Default Data Loader",
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"position": [
900,
-20
],
"parameters": {
"options": {},
"jsonData": "={{ $('Extract from File').item.json.text }}",
"jsonMode": "expressionData"
},
"typeVersion": 1
},
{
"id": "ca0c015e-fba2-4dca-b0fe-bac66681725a",
"name": "Recursive Character Text Splitter",
"type": "@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter",
"position": [
900,
100
],
"parameters": {
"options": {},
"chunkSize": 2000,
"chunkOverlap": {}
},
"typeVersion": 1
},
{
"id": "63abb6b2-b955-4e65-9c63-3211dca65613",
"name": "Extract from File",
"type": "n8n-nodes-base.extractFromFile",
"position": [
360,
-240
],
"parameters": {
"options": {},
"operation": "pdf"
},
"typeVersion": 1
},
{
"id": "be2add9c-3670-4196-8c38-82742bf4f283",
"name": "get Product Brochure",
"type": "n8n-nodes-base.httpRequest",
"position": [
180,
-240
],
"parameters": {
"url": "https://usa.yamaha.com/files/download/brochure/1/1474881/Yamaha-Powered-Loudspeakers-brochure-2024-en-web.pdf",
"options": {}
},
"typeVersion": 4.2
},
{
"id": "1ae5a311-36d7-4454-ab14-6788d1331780",
"name": "Reply To User",
"type": "n8n-nodes-base.whatsApp",
"position": [
2820,
-280
],
"parameters": {
"textBody": "={{ $json.output }}",
"operation": "send",
"phoneNumberId": "477115632141067",
"requestOptions": {},
"additionalFields": {
"previewUrl": false
},
"recipientPhoneNumber": "={{ $('WhatsApp Trigger').item.json.messages[0].from }}"
},
"credentials": {
"whatsAppApi": {
"id": "9SFJPeqrpChOkAmw",
"name": "WhatsApp account"
}
},
"typeVersion": 1
},
{
"id": "b6efba81-18b0-4378-bb91-51f39ca57f3e",
"name": "Reply To User1",
"type": "n8n-nodes-base.whatsApp",
"position": [
1760,
80
],
"parameters": {
"textBody": "=I'm unable to process non-text messages. Please send only text messages. Thanks!",
"operation": "send",
"phoneNumberId": "477115632141067",
"requestOptions": {},
"additionalFields": {
"previewUrl": false
},
"recipientPhoneNumber": "={{ $('WhatsApp Trigger').item.json.messages[0].from }}"
},
"credentials": {
"whatsAppApi": {
"id": "9SFJPeqrpChOkAmw",
"name": "WhatsApp account"
}
},
"typeVersion": 1
},
{
"id": "52decd86-ac6c-4d91-a938-86f93ec5f822",
"name": "Product Catalogue",
"type": "@n8n/n8n-nodes-langchain.vectorStoreInMemory",
"position": [
2200,
-60
],
"parameters": {
"memoryKey": "whatsapp-75"
},
"typeVersion": 1
},
{
"id": "6dd5a652-2464-4ab8-8e5f-568529299523",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-88.75,
-473.4375
],
"parameters": {
"color": 7,
"width": 640.4375,
"height": 434.6875,
"content": "## 1. Download Product Brochure PDF\n[Read more about the HTTP Request Tool](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.httprequest)\n\nImport your marketing PDF document to build your vector store. This will be used as the knowledgebase by the Sales AI Agent.\n\nFor this demonstration, we'll use the HTTP request node to import the YAMAHA POWERED LOUDSPEAKERS 2024 brochure ([Source](https://usa.yamaha.com/files/download/brochure/1/1474881/Yamaha-Powered-Loudspeakers-brochure-2024-en-web.pdf)) and an Extract from File node to extract the text contents. "
},
"typeVersion": 1
},
{
"id": "116663bc-d8d6-41a5-93dc-b219adbb2235",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
580,
-476
],
"parameters": {
"color": 7,
"width": 614.6875,
"height": 731.1875,
"content": "## 2. Create Product Brochure Vector Store\n[Read more about the In-Memory Vector Store](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.vectorstoreinmemory/)\n\nVector stores are powerful databases which serve the purpose of matching a user's questions to relevant parts of a document. By creating a vector store of our product catalog, we'll allow users to query using natural language.\n\nTo keep things simple, we'll use the **In-memory Vector Store** which comes built-in to n8n and doesn't require a separate service. For production deployments, I'd recommend replacing the in-memory vector store with either [Qdrant](https://qdrant.tech) or [Pinecone](https://pinecone.io)."
},
"typeVersion": 1
},
{
"id": "86bd5334-d735-4650-aeff-06230119d705",
"name": "Create Product Catalogue",
"type": "@n8n/n8n-nodes-langchain.vectorStoreInMemory",
"position": [
760,
-200
],
"parameters": {
"mode": "insert",
"memoryKey": "whatsapp-75",
"clearStore": true
},
"typeVersion": 1
},
{
"id": "b8078b0d-cbd7-423f-bb30-13902988be38",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
1254,
-552
],
"parameters": {
"color": 7,
"width": 546.6875,
"height": 484.1875,
"content": "## 3. Use the WhatsApp Trigger\n[Learn more about the WhatsApp Trigger](https://docs.n8n.io/integrations/builtin/trigger-nodes/n8n-nodes-base.whatsapptrigger/)\n\nThe WhatsApp Trigger allows you to receive incoming WhatsApp messages from customers. It requires a bit of setup so remember to follow the documentation carefully! Once ready however, it's quite easy to build powerful workflows which are easily accessible to users.\n\nNote that WhatsApp can send many message types such as audio and video so in this demonstration, we'll filter them out and just accept the text messages."
},
"typeVersion": 1
},
{
"id": "5bf7ed07-282b-4198-aa90-3e5ae5180404",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1640,
280
],
"parameters": {
"width": 338,
"height": 92,
"content": "### Want to handle all message types?\nCheck out my other WhatsApp template in my creator page! https://n8n.io/creators/jimleuk/"
},
"typeVersion": 1
},
{
"id": "a3661b59-25d2-446e-8462-32b4d692b69d",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
1640,
-40
],
"parameters": {
"color": 7,
"width": 337.6875,
"height": 311.1875,
"content": "### 3a. Handle Unsupported Message Types\nFor non-text messages, we'll just reply with a simple message to inform the sender."
},
"typeVersion": 1
},
{
"id": "ea3c9ee1-505a-40e7-82fe-9169bdbb80af",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
1840,
-682.5
],
"parameters": {
"color": 7,
"width": 746.6875,
"height": 929.1875,
"content": "## 4. Sales AI Agent Responds To Customers\n[Learn more about using AI Agents](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.agent/)\n\nn8n's AI agents are powerful nodes which make it incredibly easy to use state-of-the-art AI in your workflows. Not only do they have the ability to remember conversations per individual customer but also tap into resources such as our product catalogue vector store to pull factual information and data for every question.\n\nIn this demonstration, we use an AI agent which is directed to help the user navigate the product brochure. A Chat memory subnode is attached to identify and keep track of the customer session. A Vector store tool is added to allow the Agent to tap into the product catalogue knowledgebase we built earlier."
},
"typeVersion": 1
},
{
"id": "5c72df8d-bca1-4634-b1ed-61ffec8bd103",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
2620,
-560
],
"parameters": {
"color": 7,
"width": 495.4375,
"height": 484.1875,
"content": "## 5. Repond to WhatsApp User\n[Learn more about the WhatsApp Node](https://docs.n8n.io/integrations/builtin/app-nodes/n8n-nodes-base.whatsapp/)\n\nThe WhatsApp node is the go-to if you want to interact with WhatsApp users. With this node, you can send text, images, audio and video messages as well as use your WhatsApp message templates.\n\nHere, we'll keep it simple by replying with a text message which is the output of the AI agent."
},
"typeVersion": 1
},
{
"id": "48ec809f-ca0e-4052-b403-9ad7077b3fff",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
-520,
-620
],
"parameters": {
"width": 401.25,
"height": 582.6283033962263,
"content": "## Try It Out!\n\n### This n8n template builds a simple WhatsApp chabot acting as a Sales Agent. The Agent is backed by a product catalog vector store to better answer user's questions.\n\n* This template is in 2 parts: creating the product catalog vector store and building the WhatsApp AI chatbot.\n* A product brochure is imported via HTTP request node and its text contents extracted.\n* The text contents are then uploaded to the in-memory vector store to build a knowledgebase for the chatbot.\n* A WhatsApp trigger is used to capture messages from customers where non-text messages are filtered out.\n* The customer's message is sent to the AI Agent which queries the product catalogue using the vector store tool.\n* The Agent's response is sent back to the user via the WhatsApp node.\n\n### Need Help?\nJoin the [Discord](https://discord.com/invite/XPKeKXeB7d) or ask in the [Forum](https://community.n8n.io/)!"
},
"typeVersion": 1
},
{
"id": "87cf9b41-66de-49a7-aeb0-c8809191b5a0",
"name": "Handle Message Types",
"type": "n8n-nodes-base.switch",
"position": [
1560,
-280
],
"parameters": {
"rules": {
"values": [
{
"outputKey": "Supported",
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"operator": {
"type": "string",
"operation": "equals"
},
"leftValue": "={{ $json.messages[0].type }}",
"rightValue": "text"
}
]
},
"renameOutput": true
},
{
"outputKey": "Not Supported",
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "89971d8c-a386-4e77-8f6c-f491a8e84cb6",
"operator": {
"type": "string",
"operation": "notEquals"
},
"leftValue": "={{ $json.messages[0].type }}",
"rightValue": "text"
}
]
},
"renameOutput": true
}
]
},
"options": {}
},
"typeVersion": 3.2
},
{
"id": "e52f0a50-0c34-4c4a-b493-4c42ba112277",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
-80,
-20
],
"parameters": {
"color": 5,
"width": 345.10906976744184,
"height": 114.53583720930231,
"content": "### You only have to run this part once!\nRun this step to populate our product catalogue vector. Run again if you want to update the vector store with a new version."
},
"typeVersion": 1
},
{
"id": "c1a7d6d1-191e-4343-af9f-f2c9eb4ecf49",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"position": [
1260,
-40
],
"parameters": {
"color": 5,
"width": 364.6293255813954,
"height": 107.02804651162779,
"content": "### Activate your workflow to use!\nTo start using the WhatsApp chatbot, you'll need to activate the workflow. If you are self-hosting ensure WhatsApp is able to connect to your server."
},
"typeVersion": 1
},
{
"id": "a36524d0-22a6-48cc-93fe-b4571cec428a",
"name": "AI Sales Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1960,
-400
],
"parameters": {
"text": "={{ $json.messages[0].text.body }}",
"options": {
"systemMessage": "You are an assistant working for a company who sells Yamaha Powered Loudspeakers and helping the user navigate the product catalog for the year 2024. Your goal is not to facilitate a sale but if the user enquires, direct them to the appropriate website, url or contact information.\n\nDo your best to answer any questions factually. If you don't know the answer or unable to obtain the information from the datastore, then tell the user so."
},
"promptType": "define"
},
"typeVersion": 1.6
}
],
"pinData": {},
"connections": {
"AI Sales Agent": {
"main": [
[
{
"node": "Reply To User",
"type": "main",
"index": 0
}
]
]
},
"WhatsApp Trigger": {
"main": [
[
{
"node": "Handle Message Types",
"type": "main",
"index": 0
}
]
]
},
"Embeddings OpenAI": {
"ai_embedding": [
[
{
"node": "Product Catalogue",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Extract from File": {
"main": [
[
{
"node": "Create Product Catalogue",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Sales Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Product Catalogue": {
"ai_vectorStore": [
[
{
"node": "Vector Store Tool",
"type": "ai_vectorStore",
"index": 0
}
]
]
},
"Vector Store Tool": {
"ai_tool": [
[
{
"node": "AI Sales Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Embeddings OpenAI1": {
"ai_embedding": [
[
{
"node": "Create Product Catalogue",
"type": "ai_embedding",
"index": 0
}
]
]
},
"OpenAI Chat Model1": {
"ai_languageModel": [
[
{
"node": "Vector Store Tool",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Default Data Loader": {
"ai_document": [
[
{
"node": "Create Product Catalogue",
"type": "ai_document",
"index": 0
}
]
]
},
"Handle Message Types": {
"main": [
[
{
"node": "AI Sales Agent",
"type": "main",
"index": 0
}
],
[
{
"node": "Reply To User1",
"type": "main",
"index": 0
}
]
]
},
"Window Buffer Memory": {
"ai_memory": [
[
{
"node": "AI Sales Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"get Product Brochure": {
"main": [
[
{
"node": "Extract from File",
"type": "main",
"index": 0
}
]
]
},
"Recursive Character Text Splitter": {
"ai_textSplitter": [
[
{
"node": "Default Data Loader",
"type": "ai_textSplitter",
"index": 0
}
]
]
},
"When clicking ‘Test workflow’": {
"main": [
[
{
"node": "get Product Brochure",
"type": "main",
"index": 0
}
]
]
}
}
}Workflow n8n WhatsApp, intelligence artificielle, service client : pour qui est ce workflow ?
Ce workflow s'adresse aux entreprises de taille moyenne à grande, souhaitant optimiser leur service client via WhatsApp. Il est particulièrement adapté aux équipes de vente et de support technique, ainsi qu'aux professionnels du marketing digital. Un niveau technique intermédiaire est recommandé pour la mise en place et la personnalisation.
Workflow n8n WhatsApp, intelligence artificielle, service client : problème résolu
Ce workflow résout le problème de la lenteur et de l'inefficacité dans les réponses aux clients sur WhatsApp. En automatisant le processus de réponse, il élimine les frustrations liées aux délais d'attente et réduit le risque d'erreurs humaines. Les utilisateurs bénéficient d'une interaction rapide et personnalisée, ce qui améliore la satisfaction client et augmente les chances de conversion.
Workflow n8n WhatsApp, intelligence artificielle, service client : étapes du workflow
Étape 1 : Le workflow débute avec le déclencheur 'WhatsApp Trigger' qui capte les messages entrants.
- Étape 1 : Les données sont ensuite traitées par le nœud 'OpenAI Chat Model' pour générer des réponses adaptées.
- Étape 2 : Le 'Window Buffer Memory' conserve le contexte des conversations pour des interactions plus fluides.
- Étape 3 : Les informations produits sont accessibles via le 'Product Catalogue' et le 'Vector Store Tool'.
- Étape 4 : Le nœud 'Handle Message Types' dirige les requêtes vers les réponses appropriées.
- Étape 5 : Enfin, les réponses sont envoyées aux utilisateurs via le nœud 'Reply To User'.
Workflow n8n WhatsApp, intelligence artificielle, service client : guide de personnalisation
Pour personnaliser ce workflow, commencez par ajuster les paramètres du nœud 'WhatsApp Trigger' pour spécifier le numéro de téléphone et les types de messages à traiter. Modifiez les modèles d'OpenAI dans les nœuds 'OpenAI Chat Model' et 'Embeddings OpenAI' pour adapter le ton et le style des réponses. Vous pouvez également intégrer d'autres outils ou API en ajoutant des nœuds supplémentaires. Assurez-vous de sécuriser les données sensibles en configurant correctement les options de sécurité dans les nœuds de réponse. Enfin, testez le workflow régulièrement pour garantir son bon fonctionnement et son efficacité.