LOW / CODEBlueprints
generalai

AI Competitive Intelligence Monitor

Feed competitor information to Claude AI on a weekly schedule for analysis. Get automated briefings on positioning changes, new features, pricing updates, and market moves delivered to Slack and email.

Advanced~35 minn8nMake.com 640 downloads 2300 views

Setup Instructions

1. Create a Google Sheet called "Competitive Intelligence" with columns: Competitor, URL, Category (pricing/features/positioning/partnerships), Last Known State, Current State, Last Checked, Notes. Populate with your top 5-10 competitors and their key pages to monitor. 2. Set up a weekly scheduled trigger (e.g., every Monday at 7 AM). 3. Add a Google Sheets "Read" node to pull all competitor entries. 4. For each competitor, add an HTTP Request node to fetch their key pages (pricing page, features page, blog/changelog). Note: You may need to use a web scraping service or extract text from cached versions if direct fetching is blocked. 5. Add an HTTP Request node to call Claude API. Use model "claude-sonnet-4-20250514" with max_tokens: 4096. Prompt: "You are a competitive intelligence analyst. Compare the current state of these competitors against their previous known state. For each competitor, identify: 1) Pricing changes, 2) New features or removed features, 3) Positioning/messaging shifts, 4) Notable partnerships or integrations. Then provide: Overall Market Summary, Top 3 Threats, Top 3 Opportunities for our business. Previous state: [last_known_state]. Current state: [fetched_content]." 6. Add a Code node to parse Claude's analysis and format it for Slack (Block Kit) and email (HTML). 7. Add a Google Sheets "Update" node to save the current state as the new "Last Known State" for next week's comparison. 8. Add Slack and Gmail nodes to distribute the briefing. Post the full analysis to #competitive-intel on Slack and email the executive summary to leadership.
Troubleshooting
**Competitor websites block scraping:** Many sites block automated requests. Use a headless browser service, Google Cache, or manually update the "Current State" column weekly. The AI analysis is still valuable even with manually gathered data. **Analysis is too surface-level:** Add more context to your prompt — include your own product's positioning, target market, and key differentiators so Claude can make the analysis specific to your competitive situation rather than generic. **Token limits with multiple competitors:** Process competitors in batches of 2-3 rather than all at once. Then have a final Claude call that synthesizes individual analyses into the overall market summary. **Changes detected that aren't real changes:** Web page content includes dynamic elements (dates, user counts). Tell Claude to ignore: timestamps, dynamic counters, and testimonial rotations. Focus the comparison on structural content: pricing tiers, feature lists, and core messaging.

Need a custom version?

We can build a tailored automation workflow for your specific needs.

New blueprints weekly

Get notified when we publish new automation workflows.