Install
Services: Convert Your Webpages to Enriched Documents
Turn any webpage into a structured, machine-readable Enriched Document—the same schema used in 4search’s data services. Perfect for AI training, search, analytics, and content operations.
What This Service Does
Provide your URLs (or HTML/PDF files), and 4search converts each page into a high-fidelity structured object with rich metadata, entities, categories, and analysis signals. You get clean JSON/NDJSON (CSV on request) that drops straight into your pipelines—or we can index it for you in an OpenSearch/Elasticsearch cluster.
- Normalization, deduplication, canonicalization
- Entity extraction (people, organizations, places)
- Topical classification via detailed taxonomy
- Quantitative markers (money, percents, dates, durations)
- Quality metrics (readability, lengths, language, images)
Great For
- Training corpora for LLMs / RAG
- Vertical search & discovery
- Competitive intelligence & monitoring
- Editorial & research workflows
- Data marts & BI dashboards
What’s Included Per Page
Core Metadata
- Title, description, canonical URL
- Published & modified timestamps
- Site/domain, language, image/thumbnail
- Status & source identifiers (optional)
Semantic Enrichment
- Categories & tags (taxonomy-aligned)
- Entities: people, organizations, places
- Keywords (top & full)
- Money amounts, percents, durations
Analytical Signals
- Readability & grade level
- Word/syllable counts, length stats
- Index & subindex mapping aids
- Optional quality flags
Delivery Formats
- JSON / NDJSON (default)
- CSV export on request
- Direct index into OpenSearch/Elasticsearch
- S3/HTTPS drops or API retrieval
Example Enriched Document (truncated)
{
"Title": "Example Article Title",
"Description": "Short synopsis of the page ...",
"URL": "https://example.com/path/to/page",
"Date_Published": "2025-10-03 16:15:04",
"Date_Modified": "2025-10-03 16:15:04",
"Language": "en",
"Categories_Top": ["business","technology","ai"],
"People": ["Jane Smith","Alex Doe"],
"Organizations": ["Example Inc.","Research Lab"],
"Moneys_Top": ["$20 million","$3 billion"],
"Percents_Top": ["23 percent"],
"Keywords_Top": ["generative ai","model cards"],
"Tags_Top": ["ai","research","industry"],
"Image": "https://example.com/thumbnail.jpg",
"Readability_Grade": 8.5,
"Title_Word_Count": 12
} Schema aligns with the 4search Enriched Data Service for seamless downstream use.
How It Works
1) Submit
- Upload a list of URLs, a sitemap, or files (HTML/PDF)
- Set language/region hints (optional)
- Choose delivery (JSON, NDJSON, CSV, or index)
2) Convert & Enrich
- Render (with JS support as needed)
- Extract content & metadata
- Run NLP for entities, taxonomy, signals
- Normalize & deduplicate
3) QA & Delivery
- Spot checks & validation
- S3/HTTPS drop, API pull, or direct index
- Optionally, lightweight change-monitoring
Optional Add-Ons
- OCR for image-only/PDF content
- JavaScript rendering & login session handling*
- Embeddings (8-bit or float) for each document
- Change detection (diffs, deltas)
* Requires appropriate access/authorization from content owner.
API Examples
Submit URLs
POST /api/convert
Content-Type: application/json
{
"urls": ["https://example.com/a","https://example.com/b"],
"options": {
"language_hint": "en",
"include_embeddings": false,
"delivery": "ndjson"
}
} Check Job Status
GET /api/convert/status?job_id=abcd1234
{
"job_id": "abcd1234",
"submitted": "2025-10-04T16:15:00Z",
"state": "completed",
"counts": { "total": 20000, "success": 19912, "failed": 88 },
"download": "https://.../exports/abcd1234.ndjson"
} Pricing
Base Rate
$25 / 1,000 pages processed- Billed on delivered conversions
- Minimum order: 50,000 pages
- CSV export at no extra charge
Volume & Options
- Volume discounts (10M+ pages/month)
- OCR/JS rendering priced per usage
- Embeddings add-on (per 1,000 items)
- Direct indexing to your cluster (setup fee)
Need bespoke fields, custom taxonomy, or region-specific redactions? We’ll scope a tailored plan.
Terms
- Ownership: You must own the content or have rights to process it. You retain rights to your original content and to the resulting Enriched Documents.
- Redistribution by 4search: We do not redistribute your converted data without your written permission.
- Compliance: Customer is responsible for lawful submission and for honoring applicable terms/consents.
- PII/Sensitive Data: Optional redaction available upon request.
- Support: Email support included; premium SLA available.
See also our Enriched Data Access & Licensing if you need additional third-party corpora.
Ready to convert your site?
Send a URL list or sitemap, choose delivery, and we’ll handle the rest.