We built the thing we needed.

What we chose, and why

1. We identify the organization, never the individual who visited.

   Most tools in this category de-anonymize people. We resolve a visit to the organization behind it and stop there — no cookies, nothing stored on the visitor’s device, and no PII used to identify the organization.

2. We refuse to guess on shared gateways.

   A SASE or shared corporate proxy carries traffic from dozens of organizations at once. Naming one of them would be a coin flip dressed up as data, so we return nothing instead. A tool that tells you when it doesn’t know is the one you can believe when it says it does.

3. We made the score add up.

   A buying-intent score you can’t interrogate is just a vendor’s opinion with a number on it. Ours shows every factor behind it, and the factors sum to exactly the score. An AI writes the narrative around it; it never sets it.

4. We say what isn’t shipped.

   The site is checked by a test suite that fails the build if a page claims something the product doesn’t do — and that has caught us more than once. It’s cheaper than a sales call that ends in a refund.

5. We kept the full IP out of storage.

   The visitor’s IP is truncated to a /24 network block before it is stored; the full address is never retained. Session correlation is a salted hash whose secret key rotates every eight hours.

Four things, in order

Backstory identifies the organizations on your website from first-party signals, reads the pages they actually opened to work out what they’re interested in, scores how close each one is to buying with a breakdown that adds up to the number, and then finds the people to contact at that organization in the job titles you target.

Identification uses no personal data and no cookies. Contact discovery is a separate, opt-in feature that does handle personal data — business-contact details for people who work at an identified organization, from a third-party database, governed by a data processing addendum you accept first. Those two sentences are deliberately not merged.

How accurate website visitor identification is, and where it isn’t