Build a Competitor Intelligence Tracker with n8n + Google Drive + Slack

Competitor research is one of those tasks that’s always urgent but never gets done properly. Someone saves an annual report to Drive. Someone else finds a press release. A product manager screenshots an earnings call summary. By the time a strategy meeting happens, half the team hasn’t read any of it, and the other half read different documents. Nobody has a structured view of what’s actually happening with competitors.

Built a workflow that reads every competitor document the moment it lands in Drive, extracts structured intelligence, generates a strategic analysis, and pushes everything to Slack and a tracker sheet automatically.

What it does

Competitor document dropped in Google Drive folder → two AI passes (structured extraction + strategic analysis) → compiles intelligence → logs to tracker → posts full briefing to Slack

Takes about 20-25 seconds per document.

Document types it handles

  • Annual reports

  • Earnings calls

  • Press releases

  • Industry reports

  • Marketing materials

  • SEC filings

Two AI passes — structured data + strategic analysis

This workflow runs PDF Vector twice on the same document, which is the key design decision.

Pass 1 — Structured extraction:

Pulls discrete data points into a JSON schema so everything is queryable and loggable:

  • Company name, document type, report period

  • Financial metrics — revenue, growth rate, net income, market share, employee count

  • Products — name, description, status (New / Updated / Existing / Discontinued)

  • Strategic initiatives — initiative description, priority, timeline

  • Partnerships and acquisitions (as arrays)

  • Market expansion plans

  • Technology investments

  • Threats and opportunities (as separate arrays)

  • Key quotes

Pass 2 — Strategic analysis:

Uses PDF Vector’s Ask operation to generate a narrative response:

  • 2-3 sentence executive summary of the competitor’s current position

  • Top 3 strategic moves to watch

  • How this impacts your business

  • Recommended actions for your team

The structured data feeds the tracker. The strategic analysis goes into Slack where your team actually reads it.

What lands in Slack


🎯 Competitor Intelligence Update

Company: Acme Corp

Document: Annual Report

Period: FY 2024

📊 Financials: Revenue: $847.3M | Growth: 23% | Market Share: 18%

🚀 New Products: CloudSync Pro, DataVault Enterprise

📋 Strategic Initiatives:

• Expanding into APAC markets with local data residency

• Launching partner program targeting mid-market resellers

• Investing $120M in R&D for AI-native product suite

⚠️ Threats: 4 | 💡 Opportunities: 6

---

🧠 Strategic Analysis:

Acme Corp is accelerating its enterprise push with significant R&D investment and a new partner program, signaling an intent to compete more directly in the mid-market segment. The APAC expansion and local data residency features are a direct response to regulatory pressure we've seen discussed in our own roadmap. Three moves to watch: the partner program launch (could pull shared resellers), the AI product investment (18-month runway before it ships), and the APAC pricing strategy (they're likely to price aggressively to gain share).

📄 View Source Document

What lands in Google Sheets

Each row: Competitor, Document Type, Report Period, Financials, New Products, Partnerships, Acquisitions, Threats Identified (count), Opportunities (count), Source Document link, Analysis Date

Filter by Competitor to see all documents for one company. Filter by Document Type to compare annual reports across competitors side by side.

Setup

You’ll need:

  • Google Drive (a dedicated folder for competitor documents)

  • Google Sheets (free)

  • n8n instance (self-hosted — uses PDF Vector community node)

  • PDF Vector account (free tier: 100 credits/month — each document uses ~6-8 credits for two passes)

  • Slack (for strategy team notifications)

About 15-20 minutes to configure.

Download

Workflow JSON:

competitor-analysis-extractor.json

Full workflow collection:

khanhduyvt0101/workflows


Setup Guide

Step 1: Get your PDF Vector API key

Sign up at pdfvector.com — free plan works for testing. Go to API Keys and generate a key.

Step 2: Create a Google Drive folder

Create a dedicated folder called “Competitor Documents” or similar. Anyone on your team can drop documents here. Copy the folder ID from the URL.

Step 3: Create your Google Sheet

Headers in Row 1:


Competitor | Document Type | Report Period | Financials | New Products | Partnerships | Acquisitions | Threats Identified | Opportunities | Source Document | Analysis Date

Step 4: Import the workflow

Download JSON from GitHub → n8n → Import from File.

Step 5: Configure the nodes

Google Drive Trigger:

  • Connect Google Drive account (OAuth2)

  • Paste your folder ID

  • Event: File Created

Download Document:

  • Same Google Drive credential

PDF Vector - Extract Intel:

  • Add new credential (Bearer Token)

  • Paste your API key

  • Structured extraction with full JSON schema

PDF Vector - Strategic Insights:

  • Same PDF Vector credential

  • Uses Ask operation — generates the narrative strategic analysis

Compile Intelligence:

  • No config needed — combines both passes automatically

Log to Intel Database:

  • Connect Google Sheets

  • Paste your Sheet ID

Share with Strategy Team:

  • Connect Slack

  • Select your strategy or leadership channel

Step 6: Test it

Drop a competitor’s annual report or press release PDF into your Drive folder. Check Slack and your Sheet after about 30 seconds.


Accuracy

Tested on annual reports, 10-K filings, earnings call transcripts, and press releases.

  • Company name, document type, report period: ~97%

  • Financial figures (revenue, growth rates): ~94% on clearly presented tables; drops on narrative-only text

  • Product names and status: ~92% — reliable when product launches are explicitly stated

  • Strategic initiatives: ~88% — depends on how explicitly the document states priorities

  • Threats and opportunities: ~85% — best on documents that use these terms explicitly; inferred from earnings calls

  • Strategic analysis quality: strong on structured documents (annual reports, 10-Ks); more variable on press releases

Cost

Each document uses 6-8 PDF Vector credits for two passes. Free tier of 100 credits handles roughly 12-15 competitor documents per month.

Basic plan ($25/month) gives 3,000 credits — enough for a team actively tracking multiple competitors.

Customizing it

Add more competitors without changing the workflow:

Just drop documents from any company into the folder. The workflow extracts the company name automatically — no configuration needed per competitor.

Add a weekly digest:

Create a second workflow triggered every Monday that reads your Sheets tracker and posts a summary of all new competitor intelligence from the past 7 days.

Filter by company:

Add an IF node after the trigger to only process documents from specific competitors based on filename prefix (e.g., “ACME-”, “COMPETITOR2-”).

Connect to Notion:

Replace or supplement the Sheets node with a Notion node to create a database entry per document — useful if your strategy team lives in Notion.

Add web monitoring:

Pair this workflow with an n8n HTTP Request that periodically fetches competitor press release RSS feeds and routes new PDFs directly into the Drive folder.


Limitations

  • Requires self-hosted n8n (PDF Vector is a community node)

  • Uses ~6-8 credits per document due to two AI passes

  • Financial extraction accuracy depends on how clearly numbers are presented in the source document

  • Strategic analysis quality varies — stronger on formal reports than informal press releases

  • No deduplication — the same document re-uploaded creates a new row


PDF Vector n8n integration

Full workflow collection

Questions? Drop a comment.

the two-pass approach is a nice design choice — keeping structured extraction separate from the narrative analysis makes both outputs more reliable. have you run this on very large annual reports (200+ pages)? curious how PDF Vector handles chunking internally and whether the accuracy numbers hold up at that scale.

yeh Benjamin - honestly most of my testing has been on 20-80 page documents (typical quarterly reports, press releases, earnings transcripts).

For 200+ page annual reports, I’ve run maybe 3-4 through it. Here’s what I’ve seen:

Structured extraction (Pass 1): Still pretty accurate on the key sections (exec summary, financials, strategy sections). PDF Vector seems to handle the chunking internally - you don’t configure it, it just works.

Strategic analysis (Pass 2): This is where it gets interesting. On massive docs, the narrative analysis tends to focus on whatever sections have the clearest strategic language - sometimes misses stuff buried deep in appendices.

What probably happens internally: PDF Vector likely chunks the document, processes each chunk, then synthesizes. But I don’t know their exact chunking strategy.

Practical workaround: For 200+ page docs, I’ve found it works better to extract specific sections first (like just the “Strategy” or “Business Overview” section) rather than feeding the whole thing.

Have you tested large document extraction? Curious if you’ve found different results.

thanks for the detail — the section-extraction workaround makes sense as a practical fix. haven’t done much testing on 200+ page docs myself, mostly theorizing at that point. the bit about strategic analysis missing appendix content is interesting though — appendices in 10-Ks often have the most granular data (segment breakdowns, geographic splits) which can matter more than the exec summary. wonder if there’s any way to control which sections get prioritized, or whether that’s entirely internal to PDF Vector.