---
title: "Company Enrichment from Website"
canonical: https://trickest.com/library/company-enrichment-from-website
description: "Turn company homepage URLs into structured value proposition, industry, audience, and market context records in JSONL, Markdown, and HTML."
---

# Company Enrichment from Website

## Overview

A list of company URLs is not yet useful for CRM segmentation or account research because every site uses its own language. This workflow gives you one structured record per homepage with value proposition, industry, target audience, and market context fields for review and import. Fetch Homepages collects readable public text, Enrich with AI maps that evidence into the requested attributes, and Finalize Enrichment writes the machine-readable file and review documents. Add the graph to your [visual workflows](/platform/workflows) when another list needs the same field set. It normalizes company attributes from source text, rather than producing a general page summary.

## Pipeline

1. **Fetch Homepages** (`fetch-sites`, script) retrieves readable public text from each supplied company URL.
2. **Enrich with AI** (`enrich-agent`, agent) maps the fetched text into the requested company attribute fields.
3. **Finalize Enrichment** (`finalize`, script) writes the reviewed records to JSONL plus Markdown and HTML renderings.

## Inputs

- **Company homepage URLs.** The public pages to fetch and enrich. Default: replace the example URL list in Fetch Homepages with the companies in your approved scope.
- **Attribute set.** The fields Enrich with AI should populate. Default: value proposition, industry, target audience, and market context.

## Outputs

- **JSONL enrichment file.** One structured company record per homepage for CRM or research-sheet imports.
- **Markdown profile set.** The same attributes in a reviewable document, grouped by company URL.
- **HTML profile set.** A browser-ready rendering of the enriched records.

## Integrations

- HTTP
- AI

## Sample output

Illustrative output from `finalize`. The documentation domain and attribute values below show the record shape, not a real company or a completed enrichment run.

`company-enrichment.jsonl`

```jsonl
{"homepage":"https://www.example.com","value_proposition":"Illustrative placeholder for a concise offer statement","industry":"illustrative category","target_audience":"illustrative buyer group","market_context":"Derived only from the supplied homepage text"}
```

The Markdown and HTML artifacts render the same fields for review before import.

## FAQ

### Are the CRM fields configurable?

The default fields are value proposition, industry, target audience, and market context. Change the attribute set in Enrich with AI when your CRM or research sheet uses a different schema.

### Does the workflow need a company database or API?

No. It works from readable public homepage text. Join the JSONL artifact with existing CRM records after you review the source-based attributes.

### Will thin homepages produce complete records?

Pages with little readable text produce shorter, less specific records. Review those fields instead of treating unsupported detail as fact.

### Can I enrich more than one company per run?

Yes. Fetch Homepages accepts a list of URLs, and the final artifacts keep one structured record per homepage.

## Related workflows

- [Scrape and Summarize Webpages](/library/scrape-and-summarize-webpages) instead when you need a narrative recap of each page rather than normalized CRM fields.
- [AI Sample Dataset Generator](/library/ai-sample-dataset-generator) instead when no source records exist and you need synthetic rows.
- [Stock Market Analysis Digest](/library/stock-market-analysis-digest) instead when your inputs are market symbols and you need an analysis briefing rather than CRM enrichment.

---
_Markdown source of https://trickest.com/library/company-enrichment-from-website._