Skip to content

feat(site): competitive positioning rewrite — target GitHub trending#17

Merged
raphaelmansuy merged 1 commit intomainfrom
feature/site-competitive-positioning
Mar 24, 2026
Merged

feat(site): competitive positioning rewrite — target GitHub trending#17
raphaelmansuy merged 1 commit intomainfrom
feature/site-competitive-positioning

Conversation

@raphaelmansuy
Copy link
Copy Markdown
Owner

feat(site): competitive positioning rewrite — target GitHub trending

This PR rewrites the EdgeParse landing page with a data-driven competitive angle to maximize GitHub stars and developer discovery.

What changed

Hero (Hero.astro)

  • New eyebrow badge: #1 Non-ML PDF Parser
  • Competitive subtitle: ML-level accuracy without ML, 18× vs Docling, 2× vs OpenDataLoader
  • Metrics updated: 43+ pages/sec (accurate), 5 SDK languages (was 3), added WebAssembly trust badge
  • Rotating phrases now include Browser Apps, LangChain, LlamaIndex

New component: ComparisonSection.astro

  • Head-to-head score cards (EdgeParse 0.881 vs OpenDataLoader 0.844 vs Docling 0.882)
  • Full feature matrix comparing SDKs, dependencies, WASM, safety, bounding boxes
  • "Best choice" badge on EdgeParse card
  • Responsive, accessible table with badge indicators

index.mdx — 7-section landing page

  • SEO-optimised <title> and <description> targeting "fastest PDF parser zero ML"
  • QuickStart: "One Command. Instant PDF Intelligence."
  • FeatureGrid: 9 cards (up from 6) — added WASM browser support, AI safety filters, 5 SDK languages, bounding boxes for citations
  • BenchmarkSection: full 6-tool comparison chart with accurate scores
  • ComparisonSection: new component (see above)
  • AIIntegrationSection: was orphaned, now rendered
  • ShowcaseSection: 6 real-world use cases (RAG, legal, finance, academic, WASM browser, healthcare)
  • CTA: "Start Parsing PDFs in 30 Seconds — No API key. No cloud. No GPU."

Why this matters

EdgeParse has a genuinely exceptional story:

  • Matches Docling accuracy (0.881 vs 0.882) at 18× the speed
  • Beats OpenDataLoader quality (0.881 vs 0.844) at 2× the speed
  • Only PDF parser with WebAssembly (in-browser, private, offline)
  • 5 SDKs vs competitors single-language offerings

The previous site did not communicate this clearly. This PR makes the differentiation immediate and credible.

Testing

- Hero: #1 Non-ML PDF Parser eyebrow badge with competitive subtitle
- Hero: metrics 43+ pages/sec, 5 SDK languages, WebAssembly trust badge
- New ComparisonSection: head-to-head table vs OpenDataLoader/Docling/PyMuPDF
- index.mdx: SEO title/description targeting non-ML accuracy claim
- index.mdx: 7-section landing page with Benchmark, Comparison, AI Integration
- index.mdx: FeatureGrid expanded to 9 cards including WASM and AI safety
- index.mdx: ShowcaseSection with 6 real-world use cases
- index.mdx: CTA Start Parsing PDFs in 30 Seconds No API key
@raphaelmansuy raphaelmansuy merged commit f57244b into main Mar 24, 2026
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant