Google Sheets + Telegram: urgent review alerts
Your reviews are coming in. The problem is you don’t see the ones that matter until it’s too late. By the time someone spots a “package never arrived” comment (in another language), the refund is already requested and your rating has taken a hit.
This is the kind of mess support leads and e-commerce operators deal with daily. A marketing manager feels it too, because review velocity affects conversion. With Sheets Telegram alerts, urgent feedback gets surfaced fast, and routine reviews get neatly processed in the background.
This workflow pulls new reviews from Google Sheets, uses AI to translate and score them, writes the analysis back, then pings you in Telegram only when it should. You’ll see what it automates, what results to expect, and what you need to run it reliably.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: Google Sheets + Telegram: urgent review alerts
flowchart LR
subgraph sg0["Schedule Flow"]
direction LR
n0@{ icon: "mdi:play-circle", form: "rounded", label: "Schedule Trigger", pos: "b", h: 48 }
n1@{ icon: "mdi:database", form: "rounded", label: "Get Review Data", pos: "b", h: 48 }
n2@{ icon: "mdi:swap-horizontal", form: "rounded", label: "Filter Unprocessed", pos: "b", h: 48 }
n3@{ icon: "mdi:swap-vertical", form: "rounded", label: "Format Input Data", pos: "b", h: 48 }
n4@{ icon: "mdi:robot", form: "rounded", label: "AI Agent - Review Analysis", pos: "b", h: 48 }
n5@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model", pos: "b", h: 48 }
n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Parse AI Response"]
n7@{ icon: "mdi:database", form: "rounded", label: "Update Spreadsheet", pos: "b", h: 48 }
n8@{ icon: "mdi:swap-horizontal", form: "rounded", label: "Check Importance & Sentiment", pos: "b", h: 48 }
n9["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Urgent Alert (Telegram)"]
n10["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/telegram.svg' width='40' height='40' /></div><br/>Normal Notification (Telegram)"]
n1 --> n2
n0 --> n1
n3 --> n4
n5 -.-> n4
n6 --> n7
n2 --> n3
n7 --> n8
n4 --> n6
n8 --> n9
n8 --> n10
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n0 trigger
class n4 ai
class n5 aiModel
class n2,n8 decision
class n1,n7 database
class n6 code
classDef customIcon fill:none,stroke:none
class n6,n9,n10 customIcon
The Problem: Critical Reviews Get Missed (or Misread)
Most teams have a “review spreadsheet” that starts simple, then turns into a dumping ground. Someone copies reviews in, someone else skims them, and the truly urgent ones hide in plain sight. It gets worse with multilingual reviews: a short Japanese comment can read “fine” at a glance, but actually says “arrived broken” or “support never answered.” Then there’s the context switching. Checking Sheets, translating text, deciding if it’s serious, and notifying the right person isn’t hard once. It’s exhausting the 40th time.
The friction compounds. Here’s where it breaks down.
- You end up reacting late because nobody is watching the sheet every hour.
- Translation happens inconsistently, so urgency gets judged on a half-understood message.
- Manual triage becomes a bottleneck, which means “important” and “urgent” blur together.
- Reviews don’t get categorized, so patterns like “Shipping” complaints stay invisible until churn shows up.
The Solution: Translate, Score, Tag, Then Alert the Right Way
This n8n workflow turns your Google Sheet into a lightweight review monitoring system. On a schedule, it pulls rows that are marked as pending, sends each review to an AI agent (OpenAI) for translation and analysis, and then writes the structured results back into the same spreadsheet. That analysis includes an English translation, a sentiment score from -1.0 to +1.0, an urgency level (High, Medium, Low), a category tag like Shipping or Support, plus key phrases for a quick skim. After the sheet is updated, the workflow checks priority and mood. If it’s urgent, you get a Telegram alert right away. If it’s not, you still get a standard notice, so nothing disappears into a void.
The workflow starts with a scheduled run and a Google Sheets pull. AI handles the heavy thinking in the middle (translation, scoring, tagging), then the workflow routes the outcome to one of two Telegram messages. You stop babysitting your review log and start responding with context.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Say you receive 30 new reviews a day and about a third are in Japanese. Manually, even a quick process (open Sheets, translate, decide urgency, add tags, message your team) is maybe 4 minutes per review, or about 2 hours daily. With this workflow, you spend roughly 5 minutes checking Telegram and only open the sheet when something is flagged as High priority. The rest gets processed and written back automatically while you’re doing other work.
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Google Sheets to store incoming review rows.
- Telegram to receive urgent and standard notifications.
- OpenAI API key (get it from your OpenAI dashboard under API keys)
Skill level: Intermediate. You’ll connect accounts, paste credentials, and adjust a few fields in the sheet and in the AI prompt.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
A scheduled check kicks everything off. n8n runs on an interval you choose, then queries your Google Sheet for review rows.
Only pending reviews move forward. The workflow filters for items that haven’t been processed yet (based on your ProcessStatus field), so you don’t re-analyze the same review over and over.
AI translates and structures the review. The AI agent passes the text into an OpenAI chat model, which returns a translation plus a sentiment score, urgency label, category tag, and key phrases. A decoding step converts that AI output into clean fields n8n can store.
Results go back where your team already works. The workflow writes everything back to Google Sheets, then uses a priority check to decide which Telegram message to send. “High + negative” gets the urgent ping. The rest goes to a standard notice.
You can easily modify the sentiment threshold and the urgency rules based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Schedule Trigger
Set up the workflow to run on a schedule so new review records are analyzed automatically.
- Add and open Scheduled Run Start.
- In Rule, define your preferred interval (the current rule has empty interval values).
- Keep Flowpast Branding as an optional sticky note for documentation only.
Step 2: Connect Google Sheets
Configure the spreadsheet source and destination for the review records and analysis results.
- Open Retrieve Review Records and set Document to your sheet (currently
Your Google Sheet). - Set Sheet Name to
Sheet1(valuegid=0). - Open Write Back to Sheet and confirm Operation is
update. - In Write Back to Sheet, verify the Columns mapping uses expressions like
{{ $json.ReviewID }}and{{ $json.TranslatedText }}.
Step 3: Filter and Prepare Review Data
Only process reviews that haven’t been handled yet, then prepare the payload for AI analysis.
- Open Select Pending Reviews and confirm the condition checks
{{ $json.ProcessStatus }}equals an empty string. - Open Map Review Payload to map the fields used by the AI step (keep defaults if you’re passing through).
Step 4: Set Up AI Review Analysis
Configure the AI agent to translate and analyze Japanese reviews in a strict JSON format.
- Open AI Review Analyst and keep the prompt text as-is, including the review reference
{{ $('Retrieve Review Records').item.json['Keyword'] }}. - Confirm Agent is set to
conversationalAgentand Prompt Type isdefine. - Open OpenAI Chat Engine and set Temperature to
0.3to keep output consistent.
Step 5: Decode AI Output and Update the Sheet
Parse the AI response into structured fields and update the original sheet with the analysis results.
- Open Decode AI Output and keep the JavaScript Code that extracts JSON and merges it with Map Review Payload.
- In Write Back to Sheet, ensure mapping includes
{{ $json.ProcessStatus }}and{{ $json.ProcessedAt }}so rows are marked complete.
Step 6: Route by Priority and Send Telegram Alerts
Prioritize urgent reviews and notify the team through Telegram.
- Open Assess Priority & Mood and verify the conditions:
{{ $json.Importance }}equalsHighor{{ $json.SentimentScore }}is less than-0.5. - Confirm Assess Priority & Mood outputs to both Telegram Urgent Notice and Telegram Standard Notice in parallel.
- In Telegram Urgent Notice, keep the message body with expressions like
{{ $json.OriginalText }}and{{ $json.TranslatedText }}. - In Telegram Standard Notice, keep the summary message with
{{ $json.ReviewID }},{{ $json.Sentiment }},{{ $json.Importance }}, and{{ $json.Category }}.
Step 7: Test and Activate Your Workflow
Validate end-to-end processing before enabling scheduled execution.
- Click Execute Workflow and verify that Retrieve Review Records pulls rows and Select Pending Reviews filters pending items.
- Confirm AI Review Analyst returns JSON and Decode AI Output generates fields like
TranslatedTextandSentimentScore. - Check the sheet to ensure Write Back to Sheet updates the matching
ReviewIDrow withProcessStatusset toCompleted. - Verify Telegram notifications: urgent reviews appear in Telegram Urgent Notice and others in Telegram Standard Notice.
- Toggle Active to enable scheduled runs in production.
Common Gotchas
- Google Sheets access can fail if the connected Google account loses permission to the spreadsheet. If things break, check the sheet sharing settings and the Google credential in n8n first.
- If you’re processing a bigger batch of pending rows, OpenAI responses can take longer than expected. Increase your schedule interval (or add wait time) if downstream nodes start writing empty fields.
- Telegram alerts feel “noisy” when the AI prompt is generic. Add your real definitions of urgent early (refund request, safety issue, missing delivery), or you’ll be muting notifications within a week.
Frequently Asked Questions
About 30 minutes if your sheet and Telegram bot are ready.
No. You’ll mostly connect accounts and paste in your API keys. The only “technical” part is matching your Google Sheets columns to the workflow fields.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI API costs (usually a few cents per batch, depending on volume).
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, and you should. You can change the source/target language by editing the prompt inside the “AI Agent – Review Analysis” step, then adjust what counts as urgent in the “Assess Priority & Mood” checks (the default sentiment threshold is around -0.5). Common tweaks include adding new categories (Refund, Fraud, Damaged), changing the “High” definition to match your SLA, and rewriting the Telegram message so it includes order IDs or SKUs when available.
Usually it’s permissions or the wrong spreadsheet selection. Re-check the Google credential in n8n, confirm the sheet is shared with that Google account, and make sure the workflow still points to the same spreadsheet and tab (renaming tabs can break it). If it fails only sometimes, you may be hitting Google API limits during big pulls, so reduce batch size or run less frequently.
A few hundred a day is typical on a modest setup, and self-hosting scales with your server.
Often, yes, because this workflow has branching logic (urgent vs standard) plus AI parsing that benefits from more control. n8n also gives you a clean path to self-host, which matters once you’re processing reviews every hour. Zapier and Make are totally fine for simple alerting, but they can get expensive or limiting when you add filtering, formatting, and AI in the same flow. Also, this workflow writes rich fields back into Google Sheets, so your spreadsheet becomes a proper review database over time, not just a trigger source. Talk to an automation expert if you want help picking the right platform.
Once this is running, the spreadsheet stays organized and the scary reviews stop sneaking past you. You’ll see the urgent ones fast, and you can finally ignore the noise.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
