n8n workflow · Chapter 5 of 5
Manager Control Bot + Digest + Reviews
Updated · Tested on n8n 1.x · Part of the n8n Restaurant ebook
One Telegram bot gives the owner the remote control: ask about bookings and orders, get a morning digest, send review requests and broadcast offers.
How this n8n workflow works
- 1You message your Telegram bot in plain language
- 2AI reads or updates your CRM and replies
- 3Every morning you get an AI briefing
- 4On demand, recent diners get review requests
- 5Start with "broadcast" to message every customer
Who this workflow is for
- Restaurant owners/managers — check numbers, add bookings, and message customers without opening a laptop.
- Multi-outlet operators — one Telegram chat per outlet becomes a full control panel.
- Any owner-run business — the "ask anything + digest + broadcast" pattern works for salons, clinics, gyms too.
Why it matters: Most owners are on the floor, not at a computer — so data sits unused and marketing doesn't happen. A Telegram bot that answers questions instantly, nudges you every morning with priorities, and lets you broadcast an offer to your whole customer list in one line turns your phone into a command center. It closes the loop: capture (Ch1), converse (Ch2/3), retain (Ch4), and now control (Ch5).
Inside the workflow: node by node
Every one of these nodes is explained in the ebook, with each setting, prompt and expression printed so you can copy it.
- 1. Telegram Trigger — Your Command Line
- 2. Chat Trigger — The Test Entry
- 3. Command Agent — The Brain
- 4. Send Reply — Answer on Telegram
- 5. Daily 9AM — The Morning Trigger
- 6. Read Reservations (Digest) — Load the Bookings
- 7. Aggregate Leads — Bundle Into One List
- 8. Digest Agent — Write the Morning Brief
- 9. Send Digest — Deliver It
- 10. Send Review Requests — Run After Service
- 11. Read Members (Reviews) — Load All Customers
- 12. Pick Review Targets — Choose Recent Diners
- 13. Send Review WhatsApp — Ask via CPT 2
- 14. Is Broadcast? — Catch the Keyword
- 15. Read All Customers — Load Everyone
- 16. Prep Broadcast — Build One Message Per Customer
- 17. Send Broadcast — Message Everyone via CPT 2
- 18. Broadcast Confirm — Tell You It's Queued
- 19. Error Trigger — Catch Any Crash
- 20. Error Alert — Message You the Details
Before you start
- An n8n account (Cloud or self-hosted).
- A Telegram bot + your chat ID — this is both the input (your commands) and output (replies/digests).
- An OpenAI (or OpenRouter) API key — powers the command + digest agents (gpt-4o-mini).
- A Google Sheets credential — the same Restaurant CRM (Restaurant Customers, Reservations, Orders).
- Chapter 2 (CPT 2) must exist — this workflow calls it to send review WhatsApps and broadcasts.
Quick setup guide
- 1Credentials: Add Telegram, OpenAI/OpenRouter, and Google Sheets (see the video table).
- 2Import the workflow: . Reconnect every credential. Put your Sheet ID on all Google Sheets nodes and your chat ID on the Telegram nodes + Conversation Memory + Restrict-to-Chat-IDs.
- 3Point the sub-workflows: Set Send Review WhatsApp and Send Broadcast to your Chapter 2 workflow.
- 4Personalize: Edit the two agents' restaurant name, and put your Google review link in Pick Review Targets.
- 5Test: In the built-in chat, ask "aaj ki bookings?" and "aaj ka revenue"; run Send Review Requests; execute the digest; then message your bot on Telegram and try a broadcast: ... line. Then Activate.
What success looks like: You run the whole restaurant from one Telegram chat — answers on demand, a morning brief, review collection, and one-line broadcasts to your entire customer base.
Common problems and fixes
Bot ignores me
Why: Chat ID not whitelisted
Fix: Set Restrict to Chat IDs to your chat ID on the Telegram Trigger
Agent doesn't use the sheets
Why: Tools not attached / wrong Sheet ID
Fix: Attach the 5 Sheets tools; set the Sheet ID on each
"Add reservation" saves blank fields
Why: $fromAI fields unclear
Fix: Keep the field descriptions; give the command enough detail
No morning digest
Why: Schedule off or wrong time
Fix: Check Daily 9AM and that the workflow is Active
Review requests go to everyone
Why: Filter too loose
Fix: Pick Review Targets limits to served-in-3-days + ordered + not-asked
Broadcast didn't send
Why: Message didn't start with "broadcast", or CPT 2 not linked
Fix: Start the message with broadcast; point Send Broadcast at CPT 2
Image missing in broadcast
Why: No valid image URL in the message
Fix: Include a direct .jpg/.png URL in the broadcast text
Setup time
- With Claude Code30–45 minutes
Give the ebook to Claude Code and it builds the workflow for you.
- Import the JSON yourselfAbout 2 hours
Import the ready workflow JSON and connect your accounts.
- Build it by reading the ebookAbout 1 day
Follow every node and setting step by step to learn it fully.
Get this workflow
- Ready-to-import n8n workflow JSON
- Step-by-step PDF: every node and setting explained
- Copy-paste prompts and expressions — no coding
Frequently asked questions
Will this work on free n8n?
Yes. n8n itself can be free (the self-hosted Community edition) or you can use n8n Cloud. The workflow runs on both. You only pay for the outside services it uses, such as AI tokens, and many of them have free tiers to start.
How do I get the API keys (WhatsApp, OpenAI and others)?
The ebook includes a setup video for every API and account this workflow needs, so you can get each key step by step without any coding.
Do I need to know coding?
No. Every node, setting, prompt and expression is printed in the ebook, ready to copy-paste, and the workflow JSON can be imported in one click.
Can I use this Restaurants workflow for my clients?
Yes. Once it works for you, you can set it up for your own business or for clients and charge for the setup.
Bot ignores me: how do I fix it?
Chat ID not whitelisted. Fix: Set Restrict to Chat IDs to your chat ID on the Telegram Trigger.
Agent doesn't use the sheets: how do I fix it?
Tools not attached / wrong Sheet ID. Fix: Attach the 5 Sheets tools; set the Sheet ID on each.
"Add reservation" saves blank fields: how do I fix it?
$fromAI fields unclear. Fix: Keep the field descriptions; give the command enough detail.