| name | landing-page-intel-extractor |
| description | Extract competitor and customer intelligence from any company's landing page HTML. Fetches a target URL, parses the full HTML, and surfaces structured signals: named competitors, positioning claims, comparison pages, customer logos, testimonials, case studies, industries served, pricing model, integrations, and social proof. Produces a single intelligence brief ready for sales, marketing, or competitive strategy.
|
| tags | ["competitive-intel","research"] |
| graph | {"provides":["competitor-list","customer-logos","testimonials","positioning-claims","pricing-signals","integration-list"],"requires":["target-url"],"connects_to":[{"skill":"battlecard-generator","when":"Competitor names have been extracted and you need battlecards","passes":"competitor-list, positioning-claims"},{"skill":"icp-website-audit","when":"You want to evaluate the site through ICP personas","passes":"target-url, competitor-list"},{"skill":"competitor-ad-teardown","when":"You want to see what ads a discovered competitor is running","passes":"competitor domains from competitor-list"}]} |
Landing Page Intel Extractor
Turn any company's landing page into a structured intelligence brief. Fetches the HTML, extracts every competitive and customer signal on the page, and delivers a clean report you can hand to sales, marketing, or product.
Core principle: A company's landing page is a public declaration of strategy. The customers they showcase, the competitors they reference (or carefully avoid), the language they use, and the proof points they lead with — all of it is signal. This skill reads that signal systematically.
When to Use
- "What can we learn about [company] from their website?"
- "Pull competitor and customer intel from [url]"
- "Who does [company] sell to? Who do they compete with?"
- "Extract intelligence from this landing page: [url]"
- "Scrape [company]'s site for competitive signals"
- Before a sales call — quickly understand a prospect's positioning
- Building a competitive landscape — extract intel from multiple sites
Quick Start
Extract competitor and customer intelligence from https://example.com
Multiple pages:
Extract landing page intel from these URLs:
- https://example.com
- https://example.com/customers
- https://example.com/pricing
With comparison:
Extract intel from https://company-a.com and https://company-b.com, then compare their positioning.
Inputs
| Input | Required | Source |
|---|
| Target URL(s) | Yes | User provides |
| Company name | Optional | Inferred from page if not provided |
| Specific focus | Optional | e.g., "just competitors" or "just customers" |
| Client context | Optional | clients/<client>/context.md for framing |
Step-by-Step Process
Phase 1: Page Acquisition
Fetch the target URL(s) using WebFetch.
Primary page — always fetch:
WebFetch: [target_url]
Discovery pages — attempt to fetch these additional paths if they exist (non-blocking; skip 404s):
/customers or /case-studies/pricing/integrations or /partners/about/vs/ or /compare/ or /alternatives/
These secondary pages often contain the richest competitive and customer signals. Check the primary page's navigation/footer for links to these sections first, then fetch the ones that exist.
Phase 2: HTML Signal Extraction
Parse each fetched page and extract the following signal categories. Be thorough — look in headings, body text, image alt text, meta tags, schema markup, footer, and navigation.
2A: Competitor Signals
| Signal | Where to Look |
|---|
| Named competitors | Comparison pages (/vs/, /alternatives/), body text ("unlike [X]", "switch from [X]"), meta descriptions, alt text |
| Implicit competitors | Category claims ("the leading...", "#1 in..."), feature comparison tables without names, "why switch" sections |
| Positioning against | "Unlike other [category] tools", "not just another [X]", differentiation claims |
| Comparison pages | Any /vs/[competitor] or /compare/ or /alternatives/ URLs found in navigation or links |
| Migration/switch content | "Switch from [X]", "migrate from [X]", import tools mentioned |
2B: Customer Signals
| Signal | Where to Look |
|---|
| Customer logos | Logo bars/grids (img alt text, aria-labels, filenames), "trusted by" sections |
| Testimonials | Quote blocks, review widgets, named endorsements with title/company |
| Case studies | Links to /case-studies/, /customers/[name], specific metrics ("increased X by Y%") |
| Industries served | Industry pages, solution verticals, "for [industry]" sections |
| Company sizes | "Enterprise", "SMB", "startup" language, pricing tier naming, seat-based pricing thresholds |
| Named customers | Mentioned anywhere in text, case studies, press mentions |
2C: Positioning & Messaging Signals
| Signal | Where to Look |
|---|
| Hero headline | H1 tag, hero section |
| Value proposition | Subheadline, first paragraph |
| Category claim | How they define their market category |
| Key differentiators | Feature highlights, "why us" section, unique claims |
| Proof metrics | Numbers in hero/social proof ("10,000+ teams", "99.9% uptime", "50M events/day") |
| Awards/recognition | G2 badges, analyst mentions, "leader in [report]" |
2D: Product & GTM Signals
| Signal | Where to Look |
|---|
| Pricing model | Pricing page, CTA language (free trial vs. demo vs. contact sales) |
| Pricing tiers | Tier names, feature gates, price points if visible |
| Primary CTA | Hero button — reveals their conversion strategy (PLG vs. sales-led) |
| Secondary CTAs | Other conversion paths offered |
| Integrations | Integration pages, "connects with" sections, partner logos |
| Platform/tech signals | API mentions, SDK references, infrastructure claims |
| Security/compliance | SOC2, HIPAA, GDPR badges or claims |
Phase 3: Enrichment (Optional)
If the user wants deeper context, use WebSearch to:
- Verify competitor relationships — search
"[company] vs [competitor]" to confirm competitive dynamics
- Identify unlisted competitors — search
"[company] alternatives" or "[company] competitors"
- Validate customer claims — search
"[customer] + [company]" for case studies or press mentions
- Check recent news — funding, pivots, rebrand, leadership changes that add context
Only do this if the user asks for "deep" analysis or specifies enrichment. By default, stick to what's on the page.
Phase 4: Intelligence Report
Compile all extracted signals into the structured output below.
Output
Save to clients/<client-name>/intel/landing-page-intel-[company]-[YYYY-MM-DD].md
If no client context, output directly.
Report Template
# Landing Page Intelligence: [Company Name]
**URL:** [primary url]
**Extracted:** [date]
**Pages analyzed:** [list of URLs fetched]
---
## Executive Summary
[3-5 sentences. What does this company do, who do they sell to, who do they compete with, and what's their core positioning bet? Note anything surprising or notable.]
---
## Competitor Intelligence
### Named Competitors
| Competitor | How Referenced | Where Found | Context |
|-----------|---------------|-------------|---------|
| [Name] | [Comparison page / "switch from" / body mention] | [URL + section] | [Quote or context] |
### Comparison Pages Found
| URL | Competitor | Positioning Angle |
|-----|-----------|------------------|
| [/vs/competitor] | [Name] | [How they position against them] |
### Implicit Competitive Positioning
- **Category claim:** "[Exact claim from page]"
- **Differentiation language:** [How they describe what makes them different]
- **"Not" positioning:** [What they explicitly say they are NOT — often reveals who they compete with]
### Competitive Landscape Summary
[2-3 sentences. Who are the main competitors based on page signals? Where does this company position itself in the market? Any competitors conspicuously absent?]
---
## Customer Intelligence
### Customer Logos Identified
| Customer | Source | Section |
|----------|--------|---------|
| [Company name from logo/alt text] | [URL] | [e.g., "Trusted by" bar, case study] |
### Testimonials
| Quote | Person | Title | Company |
|-------|--------|-------|---------|
| "[Quote]" | [Name] | [Title] | [Company] |
### Case Studies
| Customer | Headline/Metric | URL |
|----------|----------------|-----|
| [Name] | "[Metric or headline]" | [Link if available] |
### Industries & Segments
| Industry/Vertical | Evidence |
|-------------------|----------|
| [Industry] | [Page section, solution page, case study] |
### Target Company Profile
- **Company sizes:** [Enterprise / Mid-market / SMB / Startup — based on language, pricing, logos]
[Titles mentioned, "for [role]" language, use-case framing]
[Languages offered, regional pricing, office locations]
---
| Element | Extracted Text |
|---------|---------------|
| | "[Exact text]" |
| | "[Exact text]" |
| | [How they define their category] |
| | "[Button text]" → [Action: demo/trial/signup] |
[Differentiator] — "[Supporting text from page]"
[Differentiator] — "[Supporting text from page]"
[Differentiator] — "[Supporting text from page]"
| Metric | Value | Where |
|--------|-------|-------|
| [Type: customers, users, events, uptime] | [Number] | [Section] |
[G2, analyst reports, awards]
[SOC2, HIPAA, GDPR, etc.]
[Named partners, certified integrations]
---
[Free / Freemium / Free trial / Contact sales / Usage-based / Seat-based]
[Yes/No — if yes, summarize tiers]
[Based on CTA strategy and pricing visibility]
| Integration | Category |
|-------------|----------|
| [Name] | [CRM / Analytics / Communication / etc.] |
[API availability, SDK mentions, platform claims, infrastructure details]
---
[Insight relevant to selling against or to this company]
[Key objection or positioning to be aware of]
[Messaging angle opportunity]
[Competitive positioning gap]
[Feature or integration signal]
[Market direction implied by their roadmap/positioning]
---
[Optional — full list of every extracted signal with source URL and page location, for audit purposes]
Multiple URL Mode
When given multiple URLs (e.g., a company's homepage + customers page + pricing):
- Fetch all pages in parallel
- Deduplicate signals across pages
- Note which page each signal came from
- Produce a single consolidated report
Comparison Mode
When given two or more company URLs:
- Run the full extraction for each company
- Add a Comparison section at the end:
## Head-to-Head Comparison
| Dimension | [Company A] | [Company B] |
|-----------|:-----------:|:-----------:|
| Category claim | [claim] | [claim] |
| Primary CTA | [CTA] | [CTA] |
| Customer count signal | [metric] | [metric] |
| Named customers | [count] | [count] |
| Competitors referenced | [list] | [list] |
| Pricing model | [model] | [model] |
| Key differentiator | [claim] | [claim] |
### Positioning Overlap
[Where they compete directly — same claims, same customers, same language]
### Positioning Gaps
[Where one has signals the other doesn't — potential opportunities]
Cost
Free — no paid APIs. All extraction is done via WebFetch and optional WebSearch.
| Component | Cost |
|---|
| WebFetch (page retrieval) | Free |
| WebSearch (enrichment, optional) | Free |
| Analysis | Free (LLM reasoning) |
| Total | Free |
Tools Required
- WebFetch — for retrieving page HTML
- WebSearch — for optional enrichment (competitor verification, customer validation)
- No API keys or paid tools required
Tips
- Start with the homepage, then expand. The homepage gives you positioning and social proof. Add
/customers, /pricing, and /vs/ pages for richer data.
- Check comparison pages first. If a company has
/vs/ or /alternatives/ pages, those are goldmines for competitive intel. The primary page nav or footer usually links to them.
- Logo bars are noisy. Not every logo is a current customer — some are "compatible with" or "integrates with." Look at the section heading for context.
- Alt text is your friend. Customer logos often have company names in
alt attributes or filenames even when not visible as text.
- Run on competitors too. Extract intel from your competitor's site to see who they sell to and compete with. The overlap (or lack thereof) is valuable signal.
- Pair with battlecard-generator. Once you have the competitor list, feed it into
battlecard-generator for full competitive battlecards.
- Pair with icp-website-audit. Use the extracted customer/industry signals to inform persona building, then audit the site through those personas.
- Refresh quarterly. Landing pages change. Run this periodically to catch positioning shifts, new competitors, and customer additions.
