Reply Desk: a review-reply assistant for small shops, where customers' words never leave the counter

Description and use

Reply Desk lets small shop owners quickly respond to reviews by flagging urgent complaints, classifying sentiment, and auto‑drafting replies in their voice, all while keeping data local.

Problem

Small businesses need to reply to reviews quickly but lack time; urgent complaints get buried under positive ones.

Outcome

Reply Desk enables small shop owners to prioritize urgent reviews, automatically draft polite, signed replies, and keep customer data local, all without server or API keys.

Workflow

• Set up business details once (name, type, owner name, contact).
• Paste reviews separated by blank lines.
• Keyword check flags urgent reviews (food‑safety, health, money, staff behaviour).
• If no complaint words, classify as positive; otherwise use Gemma to determine mixed or negative mood.
• Gemma drafts a reply based on mood and owner details.
• Clean‑up pass removes placeholders, markdown, and duplicate sign‑offs.
• Owner reviews, edits if needed, and posts reply.

Evidence

All of this runs on an open‑weight model, Gemma 3 1B, inside the browser tab.

Source

It puts urgent reviews first. Food‑safety, health, money and staff‑behaviour complaints are flagged "Today" and moved to the top, the moment you paste them.

Source

It reads the mood of each review: positive, mixed or negative.

Source

It drafts a reply in the owner's voice, signed with their name. Unhappy customers are pointed to the owner's own contact, like a WhatsApp number, instead of an argument in public.

Source