Decodo + Google Gemini: Glassdoor intel, ready to share
Glassdoor research sounds simple until you actually do it. The ratings are in one spot, the “why people stay” clues are buried in reviews, and you end up copy-pasting messy snippets into a doc you don’t trust.
Recruiters feel this when they’re screening five companies for a hiring plan. Analysts building a market map hit the same wall. And if you run a small team, you just need Glassdoor intel automation that turns links into consistent, comparable notes.
This n8n workflow pulls a Glassdoor company page with Decodo, then has Google Gemini structure and summarize it. You’ll see what it outputs, what you need, and how to adapt it for bulk research.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: Decodo + Google Gemini: Glassdoor intel, ready to share
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", pos: "b", h: 48 }
n3@{ icon: "mdi:robot", form: "rounded", label: "Structured Output Parser", pos: "b", h: 48 }
n4@{ icon: "mdi:robot", form: "rounded", label: "Structured Data Extractor", pos: "b", h: 48 }
n5@{ icon: "mdi:robot", form: "rounded", label: "Summarizer", pos: "b", h: 48 }
n6@{ icon: "mdi:brain", form: "rounded", label: "Google Gemini Chat Model for..", pos: "b", h: 48 }
n7@{ icon: "mdi:brain", form: "rounded", label: "Google Gemini Chat Model for..", pos: "b", h: 48 }
n8["<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/merge.svg' width='40' height='40' /></div><br/>Merge"]
n9@{ icon: "mdi:cog", form: "rounded", label: "Read/Write Files from Disk", pos: "b", h: 48 }
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/code.svg' width='40' height='40' /></div><br/>Format the Response"]
n11@{ icon: "mdi:code-braces", form: "rounded", label: "Create a Binary Response", pos: "b", h: 48 }
n8 --> n10
n2 --> n4
n2 --> n5
n5 --> n8
n10 --> n11
n1 --> n2
n11 --> n9
n3 -.-> n4
n4 --> n8
n0 --> n1
n6 -.-> n5
n7 -.-> 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 n3,n4,n5 ai
class n6,n7 aiModel
class n10,n11 code
classDef customIcon fill:none,stroke:none
class n8,n10 customIcon
The Problem: Glassdoor Research Is Painfully Inconsistent
You open a Glassdoor profile to “quickly get the vibe” and 40 minutes later you’re still scrolling. Key facts are scattered across overview modules, ratings breakdowns, and long review threads where the best insights are mixed with noise. Even when you do find what you need, it rarely fits into a clean format for comparison, so your notes end up being half quotes, half guesses. Then comes the annoying part: you repeat the same manual extraction for the next company, and the next.
It adds up fast. Here’s where it breaks down in real life.
- Copy-pasting overview details into a doc leads to missing fields, especially when you’re moving quickly.
- Reviews contain the real signal (culture, leadership, burnout), but you can’t compare unstructured paragraphs across companies.
- Manual summaries change depending on who wrote them, which means stakeholders stop trusting the research.
- Once you’re tracking more than a handful of companies, the whole process becomes a recurring weekly chore.
The Solution: Decodo Pull + Gemini Structuring + Shareable Output
This workflow turns a single Glassdoor company URL into a structured research report you can actually reuse. You start by providing two inputs: the Glassdoor company link and a geo/region value (like US or India). n8n then calls the Decodo API to fetch the raw Glassdoor content, including the company overview, ratings, reviews, and FAQs. Next, Google Gemini parses that raw payload into consistent fields, so each company comes back in the same shape. Finally, Gemini generates a clear executive-style summary, and the workflow merges everything into one final object and exports it as a JSON file you can share or feed into other systems.
The workflow starts with a manual run in n8n, so you can test it safely. After Decodo retrieves the data, Gemini handles two jobs: structured extraction (clean fields) and summarization (human-readable insights). The final output is formatted, encoded for easy handling, and written to disk as a ready-to-use report.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Say you’re comparing 10 companies for a hiring plan. Manually, if each Glassdoor profile takes about 40 minutes to pull key facts, grab 6–8 review snippets, and write a short summary, that’s roughly 7 hours of focused work (and it’s easy to make mistakes). With this workflow, you paste the URL and geo, run it, and wait for the output. Your hands-on time drops to about 5 minutes per company, so you’re spending under an hour total and letting the automation do the heavy lifting.
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Decodo for retrieving Glassdoor page data.
- Google Gemini (PaLM) to extract fields and write summaries.
- Decodo API credentials (get it from your Decodo dashboard after signup).
Skill level: Intermediate. You’ll connect credentials, edit a couple of input fields, and confirm file output settings.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
You kick it off with a run. In the current version, you click “Execute Workflow” in n8n, which is perfect for testing and one-off research. The next node sets your inputs, including the Glassdoor company_url and the geo you care about.
Decodo retrieves the raw Glassdoor content. The workflow calls the Decodo community node to fetch the company overview, rating breakdown, review content, and FAQ-like sections. It’s still raw at this stage, so it’s not something you’d want to hand to a stakeholder.
Gemini turns chaos into structure. Google Gemini runs two passes: one to extract well-defined fields (through the structured parser and extraction chain), and another to write a clear narrative summary that highlights reputation, culture themes, strengths, weaknesses, and hiring implications.
The output is packaged for sharing. n8n merges the structured fields and the summary into one object, formats it, generates a binary payload (base64-ready), and writes a JSON report to disk (for example: C:\CompanyName.json).
You can easily modify the trigger to accept a list of URLs instead of one link 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 to start the workflow and pass control to the input configuration step.
- Add the Manual Execution Start node as your trigger.
- Connect Manual Execution Start to Assign Input Parameters.
- Keep Flowpast Branding as a visual reference note (no configuration required).
Step 2: Connect Decodo and Set Input Parameters
Define the company URL and geographic context, then pull data from Decodo.
- Open Assign Input Parameters and set company_url to
https://www.glassdoor.co.in/Overview/Working-at-Decode-Technologies-EI_IE4772827.11,30.htm. - Set geo to
Indiain Assign Input Parameters. - In Decodo Data Pull, set geo to
{{ $('Assign Input Parameters').item.json.geo }}. - Set url to
{{ $('Assign Input Parameters').item.json.company_url }}in Decodo Data Pull. - Credential Required: Connect your decodoApi credentials in Decodo Data Pull.
Step 3: Set Up AI Extraction and Summarization
Run two AI branches in parallel to extract structured insights and build a summary.
- Confirm that Decodo Data Pull outputs to both Structured Insight Extract and Summary Builder in parallel.
- In Structured Insight Extract, set text to
Analyze and extract structured data {{ $json.data.results[0].content }}and keep hasOutputParser enabled. - Open Structured Parser and keep schemaType set to
manualwith the provided inputSchema JSON. - Credential Required: Connect your googlePalmApi credentials in Gemini Extraction Model and keep modelName set to
models/gemini-2.0-flash-exp. - In Summary Builder, set text to
Summarize the following content {{ $json.data.results[0].content }}and keep the manual inputSchema. - Credential Required: Connect your googlePalmApi credentials in Gemini Summary Model and keep modelName set to
models/gemini-2.0-flash-exp.
Step 4: Combine and Format the Output Payload
Merge the AI outputs and shape the final JSON payload for file output.
- Ensure Structured Insight Extract and Summary Builder both connect to Combine Results.
- Open Format Output Payload and set jsCode to
return $input.first().json.output[0];. - Verify the flow Combine Results → Format Output Payload → Generate Binary Payload.
Step 5: Configure File Output
Create a binary payload and write the structured data to disk.
- In Generate Binary Payload, keep functionCode as provided to create the base64 binary output.
- Open Disk File Writer and set operation to
write. - Set fileName to
C:\\{{ $json.overview.companyName }}.jsonin Disk File Writer. - Set dataPropertyName to
datain Disk File Writer.
Step 6: Test and Activate Your Workflow
Run a manual test to confirm the extraction and file output, then activate the workflow for production.
- Click Execute Workflow and verify that Manual Execution Start triggers the flow.
- Confirm both AI branches complete and Combine Results produces a merged output.
- Check that Disk File Writer writes a JSON file named after
{{ $json.overview.companyName }}in the target directory. - Once successful, toggle the workflow to Active for production use.
Common Gotchas
- Decodo credentials can expire or require the right plan/permissions. If things break, check your Decodo dashboard and the credential settings in n8n first.
- If you’re using Wait nodes or external processing, timing varies. Bump up the wait duration if downstream nodes fail on empty responses.
- Google Gemini prompts that are too generic will produce “nice sounding” summaries that miss what you care about. Add your evaluation criteria early (role level, department, red flags), or you will be editing outputs forever.
Frequently Asked Questions
About 30 minutes if your API keys are ready.
No. You’ll mainly connect accounts and change the input fields for company_url and geo.
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 Decodo and Google Gemini API usage costs.
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 it takes a small tweak. Replace the Manual Execution Start with a Webhook trigger or a Google Sheets trigger, then feed a list of company URLs into Assign Input Parameters. Common customizations include batch processing one sheet row at a time, saving results to Google Sheets instead of disk, and adding a “score” field for culture fit based on your criteria.
Usually it’s an expired or misconfigured Decodo API credential in n8n. Confirm the Decodo community node is installed, regenerate the key in your Decodo dashboard if needed, and make sure the request is allowed for the geo you’re passing. If it works for one company but not another, the page format may differ, so re-run with logging enabled and inspect the raw payload coming out of the Decodo Data Pull node.
On n8n Cloud Starter, you’re limited by monthly executions, so most teams keep it to a few hundred runs unless they upgrade. If you self-host, there’s no execution cap, but you’re still limited by your server resources and API rate limits from Decodo and Gemini. In practice, companies are processed one at a time per run, so scaling is mostly about batching and pacing requests.
Often, yes. This workflow benefits from multi-step parsing, merging, and file handling, and n8n is simply more comfortable with that kind of logic without turning every branch into a paid add-on. You also get a self-hosting path, which matters if you want higher volume or tighter control of data. Zapier or Make can still win for quick two-app connections, especially if your team already lives there. If you’re unsure, Talk to an automation expert and describe your volume and output needs.
Once this is running, “research” becomes a repeatable input-output habit instead of a time sink. You bring the question, the workflow brings back a clean answer.
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.
