I Launched a WhatsApp Automation Service for Small Shops — Powered by Local Open Models

Description and use

Small businesses receive instant WhatsApp replies from a local AI assistant powered by Gemma 3, eliminating cloud API costs and keeping data on-premises.

Problem

Small shops and restaurants lose customers because their owners cannot answer WhatsApp messages quickly enough.

Outcome

The service reduces response time to seconds, captures orders instantly, and keeps customer data on local hardware, eliminating per-message API costs.

Workflow

• 1. Customer sends a message on the shop’s WhatsApp number.
• 2. The local Gemma 3 model (via Ollama) processes the message and generates an instant reply.
• 3. If the query is within the FAQ set, the assistant replies directly.
• 4. If the query is an order, the assistant records it and sends an instant alert to the owner.
• 5. If the query is outside the assistant’s knowledge, it forwards the message to the owner for manual handling.
• 6. Orders are optionally logged to a spreadsheet and weekly reports are generated (Pro package).

Evidence

Small shops and restaurants in my region lose customers every day for a simple reason: nobody answers WhatsApp fast enough.

Source

an AI assistant on the shop's WhatsApp number that answers instantly, 24/7 — and it runs on open-weight models, locally.

Source

Runs on Gemma 3 via Ollama on a local machine — no per-message API cost, and customer data stays with the business

Source

Basic — $80: FAQ auto-replies + welcome message

Source