Decodo + Google Sheets: Amazon review insights
You open Amazon reviews to “quickly scan.” Then it’s 45 minutes later, you’ve read 80 opinions, and you still can’t answer the real question: what’s actually changing in customer sentiment.
This Amazon review insights automation hits product managers first, because roadmap decisions need evidence. But brand marketers and agency folks building client reports feel the same drag. You will turn scattered reviews into a clean, shareable snapshot in Google Sheets.
Below, you’ll see how the workflow pulls reviews with Decodo, summarizes themes with OpenAI, and logs the results so you can spot patterns without living in review tabs.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: Decodo + Google Sheets: Amazon review insights
flowchart LR
subgraph sg0["When clicking ‘Execute workflow’ Flow"]
direction LR
n0@{ icon: "mdi:play-circle", form: "rounded", label: "When clicking ‘Execute workf..", pos: "b", h: 48 }
n1@{ icon: "mdi:swap-vertical", form: "rounded", label: "Set the Input Fields", pos: "b", h: 48 }
n2@{ icon: "mdi:cog", form: "rounded", label: "Decodo Amazon Scraper", pos: "b", h: 48 }
n3["<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/>Extract the Reviews"]
n4@{ icon: "mdi:robot", form: "rounded", label: "Perform Review Analysis", pos: "b", h: 48 }
n5@{ icon: "mdi:code-braces", form: "rounded", label: "Create a Binary Response", pos: "b", h: 48 }
n6@{ icon: "mdi:cog", form: "rounded", label: "Write the Structured JSON to..", pos: "b", h: 48 }
n7@{ icon: "mdi:database", form: "rounded", label: "Append or update row in sheet", pos: "b", h: 48 }
n8@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Chat Model for Review..", pos: "b", h: 48 }
n3 --> n4
n1 --> n2
n2 --> n3
n2 --> n5
n4 --> n7
n5 --> n6
n0 --> n1
n8 -.-> n4
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 n8 aiModel
class n7 database
class n3,n5 code
classDef customIcon fill:none,stroke:none
class n3 customIcon
The Problem: Amazon Reviews Are Valuable, But Unusable at Scale
Amazon reviews are a goldmine until you try to use them like data. One person reads 200 reviews and walks away convinced “battery life is the issue.” Someone else reads a different 200 and swears it’s “packaging.” Nobody’s lying. You’re just sampling. And when you repeat that every week (or for multiple ASINs), you burn hours doing the same mental work again and again. Worse, manual summaries are hard to defend in a meeting because there’s no structured trail back to what customers actually said.
The friction compounds fast. Here’s where it breaks down.
- You end up copying review snippets into docs and spreadsheets, then cleaning them up like it’s 2012.
- Teams miss early signals because nobody is reading new reviews every day, especially across regions.
- Bias creeps in because you remember the extreme reviews, not the repeated themes.
- Reporting becomes a one-off effort, so trends never get tracked in the same format week to week.
The Solution: Scrape, Summarize, and Log Review Intelligence Automatically
This n8n workflow turns Amazon reviews into something you can actually work with. You start by dropping in a product URL and choosing a region (like US or India). Decodo’s Amazon Scraper pulls the latest reviews from the product page, including star ratings and review text. Then the workflow structures that messy feedback into clean JSON and asks OpenAI (GPT-4.1-mini via an OpenAI Chat Model) to summarize what customers are really saying. You get two outputs: a comprehensive summary for deeper product thinking, and an abstract summary you can paste into a weekly update without rewriting it. Finally, everything is written to a local JSON file for reference and appended to Google Sheets so you can filter, compare, and share.
The flow starts with a manual run in n8n, which makes it easy to test on one product first. From there, Decodo fetches reviews, the workflow parses and extracts themes, and Google Sheets becomes your running log. It’s review monitoring that stays consistent even when your week gets chaotic.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Say you track 3 products and do a weekly review sweep. Manually, it’s easy to spend about 45 minutes per product reading, taking notes, and writing a summary, plus another 30 minutes formatting a shareable update. That’s roughly 3–4 hours a week. With this workflow, you paste each URL once, run it, and let it write the summaries and the Google Sheets log for you. Most weeks you’re down to about 20 minutes of “check the output and share it.”
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Decodo Amazon Scraper to fetch Amazon reviews via community node.
- Google Sheets to store reviews and summaries for sharing.
- OpenAI API Key (get it from your OpenAI account dashboard)
Skill level: Intermediate. You’ll import a workflow, connect credentials, and adjust a few input fields like URL and region.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
You trigger a run with your product details. In the “Configure Input Values” step, you set the Amazon product URL, pick the geo region, and optionally name the output file.
Decodo pulls real reviews from Amazon. The Decodo Review Fetcher collects review content and related fields (like ratings) so you’re not scraping by hand or relying on guesswork.
The workflow cleans and analyzes the text. A code step parses the raw payload into structured JSON, then the AI extraction step produces two summaries: one detailed and one executive-style.
Your output is saved and logged. The workflow writes a local JSON file for reference and pushes the summaries into Google Sheets so your team can filter, compare, or build a simple dashboard.
You can easily modify the summary format to match your reporting style based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Manual Trigger
Set up the manual trigger and seed input values that drive the review fetch.
- Add and open Manual Run Trigger (no configuration needed).
- In Configure Input Values, set the following fields: amazon_url to
https://www.amazon.com/Amazon-Basics-Bluetooth-Wireless-Headphones/dp/B0BVM1PSYN, geo toIndia, and file_name toAmazon-Basics-Bluetooth-Wireless-Headphones-B0BVM1PSYN.json. - Connect Manual Run Trigger to Configure Input Values.
Step 2: Connect Decodo
Configure the Decodo integration to fetch Amazon review data using the input values.
- Open Decodo Review Fetcher and set Operation to
amazon. - Set Geo to
{{ $json.geo }}and URL to{{ $json.amazon_url }}. - Credential Required: Connect your decodoApi credentials in Decodo Review Fetcher.
- Connect Configure Input Values to Decodo Review Fetcher.
Step 3: Set Up Processing & AI Summarization
Parse the fetched review data and summarize it using the AI model.
- In Parse Review Data, keep the JavaScript code that extracts
reviewsandreviews_ai_summaryfrom the Decodo response. - Open Summarize Review Insights and set Text to
Analyze and provide me the comprehensive and abstract reviews for the following {{ $json.reviews.toJsonString() }}. - Set Schema Type to
manualand keep the provided Input Schema forcomprehensive_reviewandabstract_review. - Open OpenAI Chat Model, select the model
gpt-4.1-mini, and set Temperature to0. - Credential Required: Connect your openAiApi credentials in OpenAI Chat Model.
- Ensure OpenAI Chat Model is connected as the language model for Summarize Review Insights (credentials are added on OpenAI Chat Model, not the extractor).
Decodo Review Fetcher outputs to both Parse Review Data and Build Binary Payload in parallel.
Step 4: Configure Output Destinations
Write the raw payload to disk and append summarized insights into Google Sheets.
- In Build Binary Payload, keep the function that converts JSON into base64 binary data.
- Open Write JSON File, set Operation to
write, and set File Name to=C:\\{{ $('Configure Input Values').item.json.file_name }}. - Open Update Sheet Row, set Operation to
appendOrUpdate, and map columns: all_reviews to{{ $('Parse Review Data').item.json.reviews.toJsonString() }}and product_reviews to{{ $json.output.toJsonString() }}. - Set Document ID to your spreadsheet (replace
[YOUR_ID]) and set Sheet Name toSheet1(gid=0). - Credential Required: Connect your googleSheetsOAuth2Api credentials in Update Sheet Row.
⚠️ Common Pitfall: If Document ID remains [YOUR_ID], Update Sheet Row will fail. Replace it with your actual Google Sheet ID.
Step 5: Test and Activate Your Workflow
Run the workflow end-to-end to confirm reviews are fetched, summarized, saved to file, and recorded in Sheets.
- Click Execute Workflow on Manual Run Trigger to run a test.
- Verify that Write JSON File creates the file at
C:\Amazon-Basics-Bluetooth-Wireless-Headphones-B0BVM1PSYN.json. - Check your Google Sheet to confirm Update Sheet Row appended or updated the product_reviews and all_reviews columns.
- When the output looks correct, toggle the workflow to Active to use it in production runs.
Common Gotchas
- Decodo credentials can expire or need specific permissions. If things break, check your Decodo API key in the Decodo dashboard first, then confirm the community node is installed on your n8n instance.
- If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
- Default prompts in AI nodes are generic. Add your brand voice early or you’ll be editing outputs forever.
Frequently Asked Questions
About 30–60 minutes if your API keys and Google Sheet are ready.
No. You’ll mainly connect accounts and edit the input fields for URL and region.
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 (often a few cents per run) and your Decodo plan.
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, but you’ll want to tweak the “Configure Input Values” Set node and how you write rows in the Google Sheets step. Many teams add a simple list of ASINs/URLs, then loop through them so one run updates several rows. Another popular change is swapping the summarization prompt in the “Summarize Review Insights” AI step to match your categories (packaging, sizing, durability, shipping). If you want alerts, add Gmail or Telegram notifications when negative themes show up repeatedly.
Usually it’s an invalid or expired Decodo API key, so regenerate it in Decodo and update the credentials in n8n. If you’re on n8n Cloud, this workflow may not run at all because it uses a Decodo community node, so confirm you’re on self-hosted. Also check geo settings and request limits in your Decodo plan if you suddenly see empty results.
It depends on how many reviews Decodo returns per fetch and how much you ask the AI to summarize; on self-hosted n8n you’re mostly limited by server resources and API rate limits.
For Decodo-based scraping, n8n is usually the practical choice because you can run the community node on self-hosted and then do heavier processing in one workflow. Zapier and Make are great for simple “send this to Sheets” tasks, but they get awkward when you need code parsing, file writes, and richer AI prompting. Cost can be another factor since multi-step, high-volume runs add up quickly on per-task pricing. If you’re already in Zapier, you can still use this workflow as the “data prep” layer and then hand off summaries via Gmail or Telegram. If you want help deciding, Talk to an automation expert.
Once this is running, review monitoring stops being a recurring chore. You get clearer signals, faster updates, and a sheet you can actually build decisions on.
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.
