CORTEXFLOW · FREE FOREVER

Automation

AI Guest Messaging Assistant for Vacation Rentals

↓ Download workflow JSON

Import in n8n: Workflows → ⋯ → Import from file, then add your credentials. Free download with a free account.

How it works

Your always-on messaging desk for your vacation-rental business — routine guest questions get answered in seconds, money and safety questions wait for you:

  1. STEP 1 — Inbound guest message — the Guest Message Webhook receives each guest message (POST). Your channel-manager/PMS (Hostaway, Lodgify, Guesty…) forwards Airbnb and Booking.com threads to this webhook; email arrives through an IMAP-to-webhook bridge. Normalize Guest Message turns every payload into one shape: channel, property, guest name, message text, message ID.
  2. STEP 2 — Log to Sheets — every message is appended to the Guest Messages tab with status NEW, then Update Sheet — Status PROCESSING flips the row while the AI works.
  3. STEP 3 — AI classify + draft — AI Agent — Classify & Draft reads the message with your property context (check-in/out times, WiFi password, house rules) and returns a structured verdict via Structured Output Parser — Guest Intent: intent (wifi / checkin / checkout / houserules / amenities / parking / directions / refund / complaint / access / schedule_change / other), confidence 0–1, an auto_reply_allowed flag, a ready-to-send draft_reply, and an escalation_reason.
  4. STEP 4 — Safe auto-reply branch — IF — Safe to Auto-Reply? passes only when auto_reply_allowed = true AND confidence ≥ your threshold (default 0.85). Send Guest Reply POSTs the draft to your reply endpoint, the sheet flips to AUTO_REPLIED, and you get a Telegram summary of the guest and the reply sent.
  5. STEP 5 — Escalation branch — refunds, complaints, access codes, safety issues, early check-in/late checkout requests, or low confidence → the sheet flips to ESCALATED and 🚨 Telegram — Escalate to Host sends you the guest’s message, the AI’s read, the escalation reason, and a suggested draft so you can reply manually.
  6. STEP 6 — Acknowledge webhook — both branches merge and Respond — OK returns {"status": "ok"} to the caller.

Status flow in Sheets: NEW → PROCESSING → AUTO_REPLIED / ESCALATED

Set up steps

You’ll need:

  • n8n (self-hosted or cloud) — n8n.io
  • An OpenAI API key — platform.openai.com → API keys
  • Google Sheets OAuth2 credential — n8n Credentials → Google Sheets OAuth2 API
  • A Telegram bot — message @BotFather, send /newbot, copy the token; get your chat ID from @userinfobot
  • A channel-manager/PMS or email bridge that can POST inbound guest messages to a webhook — e.g. Hostaway, Lodgify, or Guesty for Airbnb / Booking.com threads; an IMAP-to-webhook bridge (e.g. n8n’s own IMAP trigger → webhook) for email

Steps:

  1. Import the workflow — in n8n, Workflows → Import from File → select ai-guest-messaging-vacation-rentals.json.
  2. CONFIG — Properties & Rules — open the Set node and fill in:
    • property_name — comma-separated list if you run multiple properties
    • checkin_time / checkout_time — e.g. 15:00 / 11:00
    • wifi_password — the WiFi password guests ask for most
    • house_rules_url — link to your house rules page
    • safe_intents — which intents the AI may answer alone (default wifi,checkin,checkout,houserules,amenities,parking,directions)
    • confidence_threshold — minimum AI confidence (0–1) before a message is auto-answered (0.85 is a safe start)
    • reply_api_url / reply_api_key — your PMS/channel-manager reply endpoint and key
    • telegram_chat_id and escalation_email
  3. Google Sheet — create a Google Sheet with a tab named Guest Messages and these exact headers: Message ID | Timestamp | Channel | Property | Guest Name | Message | Intent | Confidence | Status | Draft Reply | Replied At. Enter the sheet in CONFIG — Google Sheets (spreadsheet_id, sheet_name), and select your Google Sheets credential in all four Sheets nodes (Log Message — Status NEW, Update Sheet — Status PROCESSING, Update Sheet — Status AUTO_REPLIED, Update Sheet — Status ESCALATED).
  4. AI — select your OpenAI credential in OpenAI Chat Model — Guest AI (model defaults to gpt-4o-mini).
  5. Notifications — select your Telegram credential in Telegram — Notify Host (Auto-replied) and 🚨 Telegram — Escalate to Host.
  6. Activate — click Activate, copy the Guest Message Webhook production URL from the node, and point your PMS/channel-manager inbound-message webhook (or email bridge) at it. Send a test message (“What is the WiFi password?”) and check the sheet row and Telegram.

Customize it

  • Stricter auto-replies — raise confidence_threshold to 0.9+ in CONFIG — Properties & Rules so the AI answers only when very sure; borderline messages fall through to escalation.
  • Per-property answers — extend the system prompt in AI Agent — Classify & Draft with a property-specific facts block (gate codes, parking spots, quiet hours) keyed off the property field.
  • More intents — extend the parser schema’s intent list and the IF logic; add a third branch for e.g. upsell (airport transfer, late checkout fees) routed to a human draft.
  • Kill the auto-reply risk entirely — disconnect the true branch of IF — Safe to Auto-Reply? into escalation while you calibrate; reconnect it once you trust the confidence scores.
  • Audit every reply — the Draft Reply column already stores what was sent; add a weekly review by filtering rows with status AUTO_REPLIED.
  • Cheaper at scale — gpt-4o-mini is already the default; pre-filter obvious OTA auto-messages (e.g. booking confirmations) with an extra IF on message keywords before the agent runs.
  • Add a human-draft mode — clone the escalation branch, swap Telegram for a Gmail create draft to escalation_email, and send polished replies from your inbox.

Workflow nodes

This workflow is built from 17 nodes. Here's what each one does:

#NodeTypeWhat it does
1 Guest Message Webhook Webhook Entry point — receives incoming HTTP requests that trigger the workflow.
2 CONFIG — Properties & Rules Set Holds editable configuration values — fill in your own settings here before activating.
3 Normalize Guest Message Code Runs custom JavaScript to transform, filter, or compute data between steps.
4 CONFIG — Google Sheets Set Holds editable configuration values — fill in your own settings here before activating.
5 Log Message — Status NEW Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
6 Update Sheet — Status PROCESSING Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
7 AI Agent — Classify & Draft AI Agent The AI brain — reasons over the task and calls tools to get it done.
8 OpenAI Chat Model — Guest AI OpenAI Chat Model The language model powering the AI agent.
9 Structured Output Parser — Guest Intent Structured Output Parser Forces the AI to return clean, structured JSON instead of free text.
10 IF — Safe to Auto-Reply? IF Splits the flow into branches based on a condition.
11 Send Guest Reply HTTP Request Calls an external API or webhook endpoint.
12 Update Sheet — Status AUTO_REPLIED Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
13 Telegram — Notify Host (Auto-replied) Telegram Sends you a Telegram message or alert.
14 Update Sheet — Status ESCALATED Google Sheets Reads from or writes to a Google Sheet — this workflow's log and state store.
15 🚨 Telegram — Escalate to Host Telegram Sends you a Telegram message or alert.
16 Merge — Combine Branches Merge Combines the outputs of two or more branches back into one flow.
17 Respond — OK Respond to Webhook Sends the HTTP response back to whoever called the webhook.

Setup instructions

1. Add your credentials

In n8n go to Credentials → New, create each credential below, then open the named node and select it:

NodeCredential to add
Log Message — Status NEW Google Sheets OAuth2
Update Sheet — Status PROCESSING Google Sheets OAuth2
OpenAI Chat Model — Guest AI OpenAI API
Update Sheet — Status AUTO_REPLIED Google Sheets OAuth2
Telegram — Notify Host (Auto-replied) Telegram Bot API
Update Sheet — Status ESCALATED Google Sheets OAuth2
🚨 Telegram — Escalate to Host Telegram Bot API

2. Fill in your config values

Open the config node listed below and replace each placeholder with your own value before activating the workflow:

NodeFieldWhat to paste
CONFIG — Properties & Rules property_name PASTE_YOUR_PROPERTY_NAME
Your property name
CONFIG — Properties & Rules checkin_time PASTE_YOUR_CHECKIN_TIME
Your checkin time
CONFIG — Properties & Rules checkout_time PASTE_YOUR_CHECKOUT_TIME
Your checkout time
CONFIG — Properties & Rules wifi_password PASTE_YOUR_WIFI_PASSWORD
Your wifi password
CONFIG — Properties & Rules house_rules_url PASTE_YOUR_HOUSE_RULES_URL
Your house rules url
CONFIG — Properties & Rules reply_api_url PASTE_YOUR_REPLY_ENDPOINT
Your reply endpoint
CONFIG — Properties & Rules reply_api_key PASTE_YOUR_REPLY_API_KEY
Your reply api key
CONFIG — Properties & Rules telegram_chat_id PASTE_YOUR_TELEGRAM_CHAT_ID
Your telegram chat id
CONFIG — Properties & Rules escalation_email PASTE_YOUR_EMAIL
Your email
CONFIG — Google Sheets spreadsheet_id PASTE_YOUR_SPREADSHEET_ID
Your spreadsheet id

How this workflow starts

  • Webhook: Listens on path /guest-message — copy the production URL from the node after activating the workflow.

Use cases

This is an always-on messaging desk for a vacation-rental business. A Guest Message Webhook receives inbound guest messages — a channel manager or PMS like Hostaway, Lodgify, or Guesty forwards Airbnb and Booking.com threads to it, and email arrives through an IMAP-to-webhook bridge. Normalize Guest Message unifies every payload, the message is logged with status NEW, and AI Agent — Classify & Draft reads it against the property context (check-in and check-out times, WiFi password, house rules) and returns a structured verdict through Structured Output Parser — Guest Intent: intent, confidence from 0 to 1, an auto_reply_allowed flag, a ready-to-send draft reply, and an escalation reason. IF — Safe to Auto-Reply? passes only when auto-reply is allowed and confidence meets the threshold (default 0.85): the draft goes out via Send Guest Reply, the sheet flips to AUTO_REPLIED, and the host gets a Telegram summary. Anything else — refunds, complaints, access codes, safety issues, or low confidence — flips to ESCALATED and fires the Telegram — Escalate to Host alert with the guest's message, the AI's read, the reason, and a suggested draft. It exists because slow replies cost stars, and 2 AM WiFi-password questions cost sleep.

The 15-property Airbnb host losing evenings to check-in questions

A host with fifteen listings spends two hours every evening answering the same five questions: WiFi password, check-in time, parking, checkout, house rules. Those map exactly onto the default safe_intents list (wifi, checkin, checkout, houserules, amenities, parking, directions), and the answers live in CONFIG — Properties & Rules — checkin_time, checkout_time, wifi_password, house_rules_url — so the AI drafts from facts, not guesses. With the confidence threshold at 0.85, the routine questions get answered in seconds while anything ambiguous escalates. Practical tip: extend the system prompt with a per-property facts block (gate codes, parking spots, quiet hours) keyed off the property field from Normalize Guest Message, so a fifteen-property setup still answers precisely.

The co-host managing 30 listings who needs sleep

A co-host's response-time metric directly affects search ranking, but she cannot answer messages at 3 AM. The auto-reply branch is her night shift: Send Guest Reply POSTs the draft to the PMS reply endpoint, Update Sheet — Status AUTO_REPLIED records it, and Telegram — Notify Host (Auto-replied) gives her a morning summary of who got what answer. She wakes up to a log instead of a backlog. Practical tip: raise the confidence_threshold to 0.9 or higher in CONFIG — Properties & Rules — with thirty listings, she would rather escalate a borderline message than risk a wrong auto-reply at scale.

The boutique villa with strict house rules

A luxury villa owner keeps finding extra guests and late-night parties because the rules were buried in a PDF nobody read. The houserules intent plus the house_rules_url in config means every rules question gets an instant, consistent answer with the link — no more improvised replies from whoever happened to see the message. Early check-in and late checkout requests intentionally escalate, since those are revenue decisions, not FAQ answers. Practical tip: use the third-branch customization — extend the parser's intent list with upsell and route airport transfers and paid late checkouts to a human draft, turning escalations into revenue instead of chores.

The property manager who will not risk refunds going wrong

A manager with twenty doors refuses to let any software touch money conversations. The workflow agrees with her: refund and complaint intents never pass IF — Safe to Auto-Reply?, and the escalation alert includes the guest's message, the AI's interpretation, the escalation reason, and a suggested draft, so she replies in one minute instead of ten. The Draft Reply column in the sheet stores everything the AI ever sent, giving her a weekly audit trail. Practical tip: during the first two weeks, disconnect the true branch of the IF node into escalation entirely — calibrate on real messages, watch the confidence scores, and only reconnect auto-reply once the scores earn her trust.

The glamping operator with patchy connectivity

A rural glamping host is often offline for hours, which used to mean guests waited half a day for simple answers. Now the webhook, the AI agent, and the auto-reply branch run without her: routine questions resolve themselves, and the Telegram — Escalate to Host messages queue on her phone for the moment she has signal. The sheet's NEW to PROCESSING to AUTO_REPLIED or ESCALATED trail means she can reconstruct the whole day in five minutes. Practical tip: pre-filter obvious OTA auto-messages (booking confirmations) with an extra IF on message keywords before the agent runs, so the AI — and the API bill — only touches messages from real humans.

← All templates