Datasets
Buyer room
TRACKBASE.AIManage supplier listing
Trackbase — A verified crawler-access log corpus captured at the edge across the customer base. Trackbase ingests every bot hit via a Cloudflare Worker snippet, Cloudflare Logpush for larger setups, or a WordPress plugin, then auto-classifies each hit into four classes - AI crawlers (GPTBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended and more), search engines, SEO tools, and social or monitoring bots - stating each is verified rather than taken at token value. Each record carries crawler identity, the exact page read, return frequency, HTTP status code and referrer, rolled into p…
supplier index read from their site
A verified crawler-access log corpus captured at the edge across the customer base. Trackbase ingests every bot hit via a Cloudflare Worker snippet, Cloudflare Logpush for larger setups, or a WordPress plugin, then auto-classifies each hit into four classes - AI crawlers (GPTBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended and more), search engines, SEO tools, and social or monitoring bots - stating each is verified rather than taken at token value. Each record carries crawler identity, the exact page read, return frequency, HTTP status code and referrer, rolled into per-crawler timelines and top-crawled-page rankings. The byproduct across customers is an inclusion map of the live web: which pages generative engines physically fetch and which are structurally invisible to them, plus the status codes returned, which reveals whether a fetch succeeded or was blocked.
Coverage Information TechnologyAsset class EquitiesTickers GOOGL · MSFT · META
Sample, licence terms, pricing and eval results when Trackbase publishes them. Until then, discover alternatives today with a 7-day trial.
Start 7-day trial →What buyers ask about Trackbase, answered from this page.
Trackbase — A verified crawler-access log corpus captured at the edge across the customer base. Trackbase ingests every bot hit via a Cloudflare Worker snippet, Cloudflare Logpush for larger setups, or a WordPress plugin, then auto-classifies each hit into four classes - AI crawlers (GPTBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended and more), search engines, SEO tools, and social or monitoring bots - stating each is verified rather than taken at token value. Each record carries crawler identity, the exact page read, return frequency, HTTP status code and referrer, rolled into p…
Trackbase offers (Alternative, Reference) — Roughly hundreds of verified bot hits per day and a few thousand per month per installed site, on the illustrative figures shown (6,175 requests per 30 days at 47% AI share). Cross-customer volume is undisclosed and scales with install count. The 30-day window shown is a reporting view rather than a published retention guarantee, so the true retained archive is unknown and must be confirmed..
Ground truth for AI-index coverage models: which fraction of a site is reachable by generative engines and which is invisible, useful to retrieval teams estimating corpus coverage; crawler-behaviour modelling from return-cadence series; robots and policy enforcement studies, since status codes reveal whether crawlers respect blocks and whether publishers' defences actually work, which is otherwise unmeasurable; four-class bot-traffic classification training data with verified labels; SEO and answer-engine-optimisation feature sets; and publisher-side evidence for AI-access licensing disputes.
Coverage spans Europe, US; Application Software; alternative, reference; equities.
Trackbase has not added their own details yet. Not yet on file: