Claude Code for Competitive Intelligence: Build Automated Monitoring

Your competitors are watching you. They're tracking your pricing page, reading your changelog, and analyzing your job postings to figure out what you're building next. If you're not doing the same to them, you're flying blind.

Most product marketers cobble together competitive intelligence from Google Alerts, occasional website checks, and whatever the sales team remembers to forward from prospect calls. That's not a system. That's hope.

With Claude Code, you can build an automated competitive monitoring system in a single afternoon. It scrapes the pages that matter, summarizes what changed, and delivers a weekly digest straight to your inbox. I've walked dozens of PMMs through this exact project in my training sessions, and the reaction is always the same: "Wait, I can actually do this myself?"

What to Monitor (and Why)

Not everything your competitors publish is worth tracking. Focus on the signals that actually predict their strategy:

Pricing Pages

Pricing changes tell you more about a competitor's strategy than their blog posts ever will. A new enterprise tier means they're moving upmarket. A free plan means they're trying to steal your SMB customers. Price drops on specific features mean those features aren't differentiating for them. Track the full pricing page, not just the numbers. Plan names, feature lists per tier, and CTA language all carry signal.

Changelog and Release Notes

Their changelog is their product roadmap in retrospect. Track it weekly and you'll spot patterns: are they investing heavily in integrations? Doubling down on AI features? Fixing the same category of bugs over and over? That last one is a goldmine for your sales team.

Job Postings

A company's job board reveals where they're investing before any press release. Hiring 5 ML engineers? They're building AI features. Hiring a VP of Enterprise Sales? They're moving upmarket. Hiring compliance specialists? They're going after regulated industries. Job postings are leading indicators.

Homepage and Positioning

When a competitor changes their homepage headline, they're signaling a strategic shift. "The best project management tool" becoming "The AI-powered project management platform" tells you exactly where their messaging is headed. Track headline copy, hero sections, and navigation structure.

Blog and Content

Less useful for real-time intelligence, but valuable for understanding their content strategy and which buyer personas they're targeting. If they start publishing case studies in a vertical you own, that's an early warning.

Building the Scraper with Claude Code

Open your terminal and start Claude Code. Here's the prompt to kick off the project:

Create a TypeScript project called "competitive-monitor" that:

1. Reads a config file (competitors.json) listing competitors with:
   - name
   - URLs to monitor (pricing, changelog, careers, homepage)
   - check frequency (daily or weekly)

2. For each URL:
   - Fetches the page content using fetch + cheerio for HTML parsing
   - Strips navigation, footers, and boilerplate
   - Saves a clean text snapshot to a local /snapshots folder
   - Compares against the previous snapshot using a diff
   - If changes are detected, saves the diff

3. After all pages are checked, uses the Anthropic API to:
   - Summarize what changed across all competitors
   - Flag anything strategically significant
   - Rate each change as low/medium/high importance

4. Outputs a Markdown report to /reports with the date in the filename

Use cheerio for parsing, dotenv for config, and the Anthropic SDK.

Claude Code will scaffold the full project in a few minutes: config schema, scraping logic, diff engine, analysis pipeline, and report generator. If you're new to Claude Code, the setup guide gets you running fast.

Setting Up Your Competitor Config

The config file is where you define what to watch. Here's an example:

{
  "competitors": [
    {
      "name": "Competitor X",
      "urls": [
        { "label": "pricing", "url": "https://competitorx.com/pricing", "frequency": "daily" },
        { "label": "changelog", "url": "https://competitorx.com/changelog", "frequency": "daily" },
        { "label": "careers", "url": "https://competitorx.com/careers", "frequency": "weekly" },
        { "label": "homepage", "url": "https://competitorx.com", "frequency": "weekly" }
      ]
    },
    {
      "name": "Competitor Y",
      "urls": [
        { "label": "pricing", "url": "https://competitory.io/pricing", "frequency": "daily" },
        { "label": "changelog", "url": "https://competitory.io/releases", "frequency": "daily" },
        { "label": "careers", "url": "https://competitory.io/jobs", "frequency": "weekly" }
      ]
    }
  ]
}

Start with 2 or 3 competitors and 3 to 4 URLs each. You can always add more later. The point is to get something running today, not to boil the ocean.

What the Output Looks Like

After a week of snapshots, your report looks something like this:

# Competitive Intelligence Report: Week of Feb 24, 2026

## 🔴 High Priority

### Competitor X, Pricing Change
Competitor X dropped their Pro tier from $79/mo to $59/mo and added a new
"Business" tier at $129/mo with SSO and audit logs. This suggests they're
splitting their mid-market and enterprise plays. The Pro tier is now
priced directly against our Growth plan.

**Action items:**
- Update competitive battlecard with new pricing
- Alert sales team about the Pro tier price drop
- Review our Growth plan positioning

## 🟡 Medium Priority

### Competitor Y: New Feature (Changelog)
Shipped "AI Meeting Summaries" as a beta feature on their Business plan.
Third AI feature released in 6 weeks. They're clearly investing heavily
in AI capabilities as a differentiator.

### Competitor X: Job Postings
Posted 3 new roles: Senior ML Engineer, AI Product Manager, and
Healthcare Solutions Architect. The healthcare hire is new; they may
be entering the healthcare vertical.

## 🟢 Low Priority

### Competitor Y: Homepage
Updated hero copy from "Simple project management" to "AI-powered
project management for modern teams." Navigation unchanged.

That report took zero manual effort. The scraper ran, the diffs were computed, and Claude's API wrote the analysis. Your job is to read it and decide what to act on.

Running It on a Schedule

A manual script is good. An automated one is a weapon. Ask Claude Code:

Add a cron-friendly mode that:
1. Runs all checks based on their configured frequency
2. Only generates a report if changes were detected
3. Sends the report via email using Resend (store API key in .env)
4. Keeps a running log of all detected changes in a changes.json file
   so I can look at trends over time

On a Mac, you can schedule this with a cron job or launchd. On a server, use cron or a simple GitHub Action. The daily checks take about 30 seconds to run and cost pennies in API calls.

For the email delivery, Resend gives you 100 free emails per day. More than enough for a daily competitive digest.

Handling JavaScript-Heavy Pages

Some competitor pages render with JavaScript, which means a simple HTTP fetch gets you an empty shell. If you hit this, tell Claude Code:

Some of these pages are SPAs that render with JavaScript. Add a fallback
mode using Puppeteer that:
1. Launches a headless browser
2. Waits for the page to fully render
3. Extracts the rendered HTML
4. Falls back to Puppeteer automatically when cheerio returns
   suspiciously little content (under 200 characters of text)

Claude Code will bolt on Puppeteer as a fallback without rewriting the existing cheerio logic. The two approaches coexist: fast HTTP fetch for static pages, headless browser for SPAs.

Adding Trend Analysis

After a month of data, the real insights emerge. Ask Claude Code to add a monthly trend report:

Using the changes.json log, generate a monthly trend report that shows:
- Which competitors are shipping fastest (changelog update frequency)
- Pricing movement direction (up, down, restructured)
- Hiring velocity by department
- Messaging shifts over time
- A "threat assessment" ranking competitors by activity level

Now you're not just reacting to individual changes. You're seeing the arc of each competitor's strategy. A competitor that shipped 12 features in a month and hired 8 engineers is investing aggressively. One that changed its pricing twice and updated its homepage copy three times is struggling to find positioning. These patterns are invisible in one-off checks but obvious in trend data.

What This Costs

Let's break it down:

• Claude Code subscription: $20/month (Pro plan)
• Anthropic API calls for analysis: roughly $2 to $5/month depending on volume
• Resend for email delivery: free tier
• Your time to build it: one afternoon
• Your time to maintain it: maybe 5 minutes/month to add new URLs

Compare that to Klue, Crayon, or Kompyte, which run $20K to $60K per year. Those platforms are powerful, but if your company won't fund a competitive intelligence tool, you can build 80% of the functionality yourself for the cost of a lunch.

The Bigger Picture

This project illustrates something I keep seeing in my ClaudeFluent training: the tools PMMs have been requesting for years are now buildable in hours. Win/loss analysis, competitive monitoring, content audit tools, sales enablement dashboards. The bottleneck was never your ideas. It was access to engineering resources. Claude Code removes that bottleneck.

The PMMs who pick up this skill now will have a compounding advantage. While everyone else waits for tools, you'll build them. While everyone else reads competitor blogs manually, you'll get automated digests. That gap widens every month.

Get Started

If you want to build projects like this with guided instruction, ClaudeFluent runs live cohorts where you go from zero to building real tools in a few hours. No coding background required. Check the homepage for the next session.