T2 Industries

Pre-launch · T2Bot identity published

The web, indexed fresh.
Sold as data.

T2 Industries crawls the public web, keeps an index that is re-fetched on a schedule tied to how fast each page actually changes, and sells search, enrichment, categorisation and change data built from it. Every record carries the time it was fetched and a content hash, so you always know how fresh it is.

See the products About T2Bot

  • Public pages only
  • No visitor data, ever
  • Never AI training
0hrobots.txt of every known host re-read
dailyre-visit for news and fast-changing pages
0M+pages in the body tier, refreshed monthly
0bytes of visitor data collected. Ever.

Fresh by design

Data that is as current as the web it came from

Most web data is a snapshot that was already stale when it shipped. T2 is built around re-fetching: every page has a cadence tied to its own change history, and every record tells you exactly when it was last confirmed.

24 h · robots.txtdaily · fast pagesweekly · headmonthly · body

Cadence follows change

Pages that change often are re-fetched daily. Stable pages roughly monthly. The head of the web, about a million of the most-referenced sites, weekly at least. Nothing sits untouched for a year.

Provenance on every record

Each result and each enrichment carries fetched_at and a content_hash. You can see freshness per record, not per dataset, and prove where a fact came from.

Cheap to stay current

T2Bot sends conditional requests, so an unchanged page is confirmed with a 304 and no body. That lets us re-check far more often than a crawler that downloads everything every time.

Change is a product

New domains, domains that went dark, pages whose content changed, new URLs per host: the T2 Changes feed tells your systems what to re-fetch instead of guessing.

robots.txtevery 24 h
Fast-changing pagesdaily
Head · ~1M sitesweekly
Body · ~100M pagesmonthly
Tailquarterly or on a change signal

Products

Five ways to buy the index

We sell facts, extractions, scores and retrieval. Never bulk page text for model training.

T2 Search API

Full-text search with the clean page text in the response, deduplicated, with provenance.

Applications, agents, research tools

T2 Enrich

One call per domain: company, description, published contacts, tech stack, category, size signals.

CRM enrichment, lead qualification, due diligence

T2 Categories

Domain-to-category database: IAB taxonomy plus security categories, daily file or lookup API.

DNS filters, firewalls, brand safety

T2 Changes

Feeds of new domains, dark domains, changed pages and new URLs per host.

Security, monitoring, retrieval systems

T2 Opt-Out Registry

Which domains allow or block which crawler tokens, tracked over time. Metadata only.

Compliance teams

Built for machines

A web result and the text behind it, in one call

No scraping, no headless browsers, no rate-limit roulette. The Search API returns ranked results with the extracted main text, language, canonical URL, fetch time and content hash, so an agent or an application can cite and trust what it reads.

  • BM25 relevance over title, headings and body
  • Host quality, freshness and sitemap lastmod in the ranking
  • No click data, no browsing data, no user-behaviour signals. We have none.

Read the docs Pricing

t2 search api
$ curl -s https://api.t2ind.com/v1/search?q="rack+cooling+standards" \
     -H "Authorization: Bearer $T2_KEY"
{
  "query": "rack cooling standards",
  "took_ms": 41,
  "results": [{
    "url": "https://example.org/guides/rack-cooling",
    "title": "Rack cooling: ASHRAE classes explained",
    "lang": "en",
    "fetched_at": "2026-10-21T04:12:09Z",
    "content_hash": "sha256:9f2c…e1a0",
    "text": "Data-centre racks are rated against ASHRAE A1–A4 thermal envelopes…"
  }]
}

How it works

Crawl. Index. Products.

  1. 01

    Crawl

    T2Bot fetches publicly served HTML, at most one request every five seconds per host, from published IP addresses with matching reverse DNS. robots.txt is honoured exactly.

  2. 02

    Index

    Each page is reduced to its facts: URL, title, headings, main text, links, structured data, language and a content hash, stamped with the fetch time.

  3. 03

    Products

    Search, enrichment, categories, change feeds and the opt-out registry are derived from the index and delivered by API or daily file.

How the T2 Index is built

For site owners

An honest crawler

Running a website? T2Bot identifies itself, honours robots.txt and publishes its IP list. It never logs in, never submits forms, never sets a cookie, and never feeds AI model training. Details and opt-out on the T2Bot page.

T2Bot identity page Opt out

Ready for data that is actually current?

Pre-launch pricing is on the site. Tell us what you are building and we will come back with a plan.

data@t2ind.com