Skip to content
Back to Blog
Tutorial

WhatsApp Bot in a Weekend with n8n

Valmir Hazeri March 19, 2026 7 min read
WhatsApp Bot in a Weekend with n8n

WhatsApp has 2 billion monthly active users worldwide, and in the DACH region it is the dominant messaging platform — used by over 80% of smartphone owners in Germany, Austria, and Switzerland (Statista, 2024). For businesses, the WhatsApp Business API unlocks automated conversations at a scale that email and web chat cannot match. Juniper Research projects that chatbot interactions on messaging apps will reach 134 billion annually by 2025, with retail and service businesses leading adoption.

The good news: you do not need a development team or a six-figure budget to build a capable WhatsApp bot. With n8n (an open-source workflow automation platform), the WhatsApp Business API, and an LLM for message classification, a functional bot handling order lookups, booking confirmations, and FAQ routing can be built in a weekend. This guide walks you through the setup from a business perspective — not a developer tutorial, but a practical roadmap with real costs, timelines, and compliance requirements for EU businesses.

Setting Up the WhatsApp Business API

The WhatsApp Business API is not the same as the free WhatsApp Business App you may already use on your phone. The API is designed for automated, scalable messaging and requires approval through Meta Business Suite. Here is what the setup process looks like:

Step 1: Create a Meta Business Account. If you already run Facebook or Instagram ads, you likely have one. Navigate to business.facebook.com, verify your business with a phone number and business documents. For German businesses, your Handelsregister-Auszug or Gewerbeanmeldung works. Verification typically takes 2–5 business days.

Step 2: Choose a Business Solution Provider (BSP). Meta does not give direct API access to most businesses. You work through a BSP — the most common options are Twilio (starting at $0.005/message plus carrier fees), 360dialog (popular in Europe, from €49/month flat for one number), and MessageBird. For DACH businesses, 360dialog is often the best fit because they host data within the EU and offer German-language support. Twilio is the most documented option and integrates natively with n8n.

Step 3: Register your phone number. You need a dedicated phone number for the bot — it cannot be the same number already registered to WhatsApp. A German mobile or landline number works. The BSP connects this number to the WhatsApp Business API. Once registered, your business appears with a verified badge and business profile (logo, description, address, website).

Step 4: Set up message templates. WhatsApp requires pre-approved templates for outbound messages (messages you initiate). These must be submitted through Meta Business Suite and approved before use — approval takes 1–24 hours. Inbound messages (customer-initiated) do not require templates, which is where your bot handles most interactions. Template examples: order confirmation, appointment reminder, shipping notification.

Cost overview: The WhatsApp Business API charges per conversation (a 24-hour window). In Germany, a service conversation costs approximately €0.0311 and a marketing conversation costs €0.0679 (Meta pricing, 2024). With 360dialog at €49/month plus per-conversation fees, a bot handling 2,000 service conversations per month costs roughly €111/month in total platform fees. Twilio pricing is comparable but billed per-message rather than per-conversation.

Building the Bot Logic in n8n

n8n is an open-source, self-hostable workflow automation tool — think of it as a visual programming environment where you connect triggers, logic nodes, and actions without writing code. The cloud-hosted version starts at €20/month (Starter plan, 2,500 executions). Self-hosting on a €5/month VPS (Hetzner, Contabo) is free for the software itself. For a WhatsApp bot, n8n is the orchestration layer: it receives messages from the WhatsApp API, processes them, and sends responses back.

Core workflow architecture: Your n8n workflow starts with a Webhook node that receives incoming WhatsApp messages from your BSP. Every time a customer sends a message, the BSP forwards it to your n8n webhook URL as a JSON payload containing the sender number, message text, and metadata. From there, the workflow branches based on message intent.

Use Case 1 — Order status lookups: The customer sends a message like "Where is my order?" or provides an order number. An n8n IF node checks whether the message contains an order number pattern (e.g., #ORD-12345). If it does, an HTTP Request node queries your shop system API (Shopify, WooCommerce, or your ERP) with that order number and returns the current status. The response is formatted into a WhatsApp message: "Your order #ORD-12345 shipped on March 15 via DHL. Tracking: [link]." For a typical Shopify store, this workflow takes 30 minutes to build in n8n.

Use Case 2 — Booking confirmations: Customers send messages requesting appointments. The workflow captures the preferred date and service type, checks availability against your Google Calendar or booking system (Calendly, Cal.com) via API, and either confirms the slot or offers alternatives. A confirmation message template is sent with the booking details. For service businesses — salons, clinics, consulting firms — this eliminates 60–80% of scheduling phone calls.

Use Case 3 — FAQ routing with LLM classification: This is where the bot becomes intelligent. Instead of building rigid keyword matching for every possible customer question, you send the incoming message to an LLM (GPT-4o via OpenAI API at ~$0.0025/request, or Claude via Anthropic API) with a classification prompt. The prompt instructs the model to categorize the message into predefined intents: order_status, booking, return_request, pricing_question, human_handoff. The LLM returns the category, and n8n routes the conversation to the appropriate sub-workflow. For questions the bot cannot handle, it routes to a human agent with the full conversation context. This classification approach handles 85–90% of inbound messages without any manual keyword mapping — and adapts to how customers actually phrase their questions in German, Austrian dialect, or Swiss German.

GDPR Compliance and Going Live

Running an automated WhatsApp bot for EU customers triggers several DSGVO (GDPR) obligations. Ignoring them is not an option — fines can reach €20 million or 4% of annual turnover. Here is what you need to get right before going live:

Consent and opt-in: Customers must actively opt in to receive WhatsApp messages from your business. A pre-checked checkbox on your website does not count. Best practice: a separate opt-in step where the customer sends the first message to your WhatsApp number (customer-initiated conversations satisfy consent requirements) or explicitly agrees via a double opt-in form. Document every opt-in with a timestamp and the specific consent text shown.

Data processing agreement (Auftragsverarbeitungsvertrag, AVV): You need a signed AVV with every third party that processes customer data — your BSP (Twilio, 360dialog), your hosting provider if you self-host n8n, and OpenAI/Anthropic if you use their LLMs for message classification. Meta provides a standard data processing agreement for the WhatsApp Business API. 360dialog offers an EU-hosted solution with a GDPR-compliant AVV. For OpenAI, note that API usage (not ChatGPT) has a separate data processing addendum — API inputs are not used for model training under the current terms, which is the configuration you need.

Data minimization and retention: Store only the data you need, and define retention periods. Customer phone numbers and conversation logs should have a clear deletion schedule — 90 days is a common baseline for support conversations. In n8n, configure your workflows to avoid logging full message content in execution history for production environments. If you use a database to store conversation state, add automated cleanup jobs.

Privacy policy update: Your Datenschutzerklaerung must explicitly mention WhatsApp Business as a communication channel, name the BSP as a processor, explain what data is collected (phone number, message content, timestamps), state the legal basis (consent per Art. 6(1)(a) DSGVO or legitimate interest per Art. 6(1)(f) for existing customer service), and list data transfers to third countries if applicable (e.g., if using US-based OpenAI, reference the EU-U.S. Data Privacy Framework).

Going live checklist:

  • WhatsApp Business API approved and number verified
  • n8n workflows tested with 50+ sample messages across all intent categories
  • LLM classification prompt tuned for German-language input (test with informal and formal address, dialect variations)
  • Fallback to human agent working and monitored
  • All AVVs signed with BSP, hosting, and AI providers
  • Privacy policy updated and published
  • Opt-in mechanism in place and documented
  • Message templates approved by Meta
  • Monitoring dashboard set up in n8n to track failed executions

Total weekend budget: WhatsApp BSP setup (€49–€99/month), n8n cloud or self-hosted (€0–€20/month), LLM API costs (~€5–€15/month for 2,000 classified messages), domain and VPS if self-hosting (€5–€10/month). All in: €60–€145/month for a fully functional, GDPR-compliant WhatsApp bot handling order tracking, bookings, and intelligent FAQ routing.

Key Takeaways

  • The WhatsApp Business API with n8n and an LLM for intent classification can handle 85–90% of inbound customer messages automatically — order lookups, booking confirmations, and FAQ routing — at a total cost of €60–€145/month.
  • GDPR compliance is non-negotiable for EU WhatsApp bots: you need documented opt-in, signed data processing agreements with every third-party provider, a defined data retention schedule, and an updated privacy policy.
  • Start with one use case (order status is the fastest to build — 30 minutes in n8n), validate it with 50 real customer messages, then add booking and FAQ workflows incrementally over the weekend.

Conclusion

A WhatsApp Business bot is not a novelty project — it is a direct channel to where your customers already spend their time. In the DACH region, WhatsApp is the default messaging tool for over 60 million users. A bot that answers order questions in 3 seconds instead of a 48-hour email turnaround is not a nice-to-have; it is a measurable improvement in customer experience and operational efficiency. The technical barrier has dropped to near zero: n8n provides the visual workflow builder, BSPs handle the API complexity, and LLMs handle the language understanding that would have required a dedicated NLP team five years ago.

The weekend timeline is realistic if you scope it correctly. Saturday: API setup, n8n installation, first order-status workflow. Sunday: booking confirmation flow, LLM classification for FAQ routing, GDPR documentation. Monday: test with real customers, monitor, and iterate. d2b builds production-grade WhatsApp automation for businesses in the DACH region — if you want a bot that handles more than the basics, or need to integrate with your existing ERP and CRM systems, the contact form is below.

Valmir Hazeri
Valmir Hazeri

Founder of d2b — building private AI automation and Gen-AI solutions for businesses across Europe.

Ready to Automate Your Business?

Let's discuss how AI automation can transform your daily operations and free up your time for what matters most.

Book a Free Audit