Turn WhatsApp into a smart sales chat that answers product questions from your brochure. It supports sales teams who need fast and accurate replies without digging through PDFs. Use it to guide buyers and move them toward a decision.

A one time setup downloads your brochure PDF, pulls the text, splits it into small parts, and builds a searchable catalog using OpenAI embeddings. Incoming messages land through the WhatsApp trigger and non text messages are filtered with a simple reply. The AI agent uses a chat model and a memory key tied to each phone number, so conversations stay in context. When a question arrives, the agent looks up facts in the vector store tool and returns a clear answer over WhatsApp.

You need a WhatsApp Business account and an OpenAI API key. Activate the workflow and test with your own PDF or keep the sample file. Expect response time to drop from minutes to seconds, with capacity to handle many chats across retail, events, or B2B catalogs.

• WhatsApp trigger captures incoming customer messages in real time
• Message type filter replies to non text posts with a clear instruction
• One time catalog build downloads a PDF and extracts the text
• Text is split into chunks for better search and matching
• OpenAI embeddings create a searchable catalog stored in memory
• AI agent uses an OpenAI chat model with session memory per phone number
• Vector store tool finds relevant brochure sections for each question
• WhatsApp node sends the agent reply back to the customer

• Reduce manual brochure lookups from 5 minutes to 15 seconds per question
• Automate 80 percent of common product questions
• Improve answer accuracy by 30 percent using a vector store built from your PDF
• Handle 10 times more simultaneous chats without extra staff
• Connect WhatsApp and AI in one flow without extra tools
• Keep replies consistent by using one trusted product source

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with WhatsApp and OpenAI. See the Tools Required section above for links to create accounts with these services.
3. OpenAI credentials: In the n8n credentials manager, create a new OpenAI credential. Paste your API key from your OpenAI account page. Name the credential clearly, for example OpenAI Sales Chat.
4. WhatsApp credentials: Double click the WhatsApp Trigger node, choose Create new credential, then follow the on screen steps to connect your WhatsApp Business account. Name it clearly, for example WhatsApp Sales Bot.
5. Update your sending number: Open both Reply To User nodes and replace the phoneNumberId field with your own number ID from the WhatsApp Business dashboard.
6. Webhook connection: In the WhatsApp Trigger node, copy the webhook URL if shown and follow the Meta setup guide to subscribe to messages for your number. If unsure, use the credential wizard and follow the prompts.
7. Catalog source: In the HTTP Request node named get Product Brochure, keep the sample URL or paste your own brochure link. Make sure the PDF is publicly reachable.
8. Build the catalog: Click Execute on the manual path using When clicking Test workflow. Wait for Extract from File, Text Splitter, and the Create Product Catalogue nodes to finish without errors.
9. Agent setup check: Ensure the AI Sales Agent node is connected to the OpenAI Chat Model, the memory node, and the Vector Store Tool. No changes are needed to default prompts for basic use.
10. Activate the workflow: Turn it on in n8n. Send a text message to your WhatsApp business number and ask a product question. You should receive a direct reply. Send an image to see the non text notice.
11. Troubleshooting: If answers are empty, confirm the PDF link works and the Extract from File node returns text. If replies are generic, lower the chunk size in the Text Splitter or add more product pages. If WhatsApp does not deliver, recheck your number ID and app permissions.