About Semantil
Company and market research on our own index of websites.
What we do
Semantil researches markets using its own index of websites. It finds relevant companies, counts and groups them, explains why they match, and tells you what the data cannot confirm.
It works in the browser through the Agent and the Catalog, through a REST API and as an MCP server for AI agents.
Where the data comes from
Semantil runs its own crawl, its own technology detection and its own search engine. The data is not bought from or resold by other providers.
Every record carries the date it was last checked. Contact details carry the page they came from. Coverage is stated for each signal, and a missing value means not detected, not absent.
How the index, the engine and the estimates work, and what we do not collect: Technology & data.
What a record includes
Domain, title, description, country, language, the year we first saw the website, business market and business segment.
Detected website technologies and, for stores, the ecommerce platform.
Product categories, payment methods, carriers, countries sold to and an estimated size band, where our crawl detects them.
A checksum-valid VAT number, one company email and one phone number published on the company's own website, each with its source page and date. No personal mailboxes and no named people.
Every field and its meaning: Data dictionary.
Semantil and ShopRank
Semantil is the Agent and workspace for researching companies and maintaining lists. ShopRank provides exhaustive e-commerce depth, historical datasets and separately licensed feeds.
Embedded use, redistribution, resale and sublicensing of the data require an explicit additional licence.
Built from real infrastructure
The infrastructure behind Semantil grew out of years of crawling, classification, technology detection, and search work, running since 2017. Every public product surface is backed by our own observation of the web.