Draft — not yet in force. This text is a plain-English draft that has not been approved. Nothing on this page binds anyone until it is.

Data processing agreement

This agreement applies when you use Linktrail to record information about people. It is the contract GDPR (the EU’s data protection law) Article 28 requires between a business that decides how personal data is used (the controller, you) and a business that handles it on its behalf (the processor, us). It forms part of our terms.

1. The parties

Controller: you, the business that holds the Linktrail account.

Processor: to be supplied (to be supplied), company number to be supplied, VAT number to be supplied, registered at to be supplied. Contact for data protection: to be supplied.

2. What we do, and for how long

We record and report email opens, link clicks and website activity for you, for as long as your account exists, and we process personal data only for that purpose. We do not use it for our own purposes.

3. Whose data, and which data

The people the data is about are:

  • people you send tracked links or emails to;
  • visitors to your websites that run our script, including visitors we cannot name, such as a stranger who clicks a link you shared publicly.

The kinds of data are:

  • the reference you give us for a person. We do not require a name or an email address, but a contact also has a free-text label and optional metadata (up to 16 KB), and whatever you put there is stored as you give it;
  • email opens, link clicks and page views, with their times, and for page views the page path and active time;
  • the IP address behind a link click or an email open, and, for link clicks only, the country derived from it;
  • for each email open or link click, a connection kind sorted from that address when it arrives (data centre, home or mobile, mail-provider proxy, scanner, or unknown), and a reader label, a number per email (r1, r2, ...) assigned in order and never computed from the address, kept as long as the email’s record exists and removed when the person is erased;
  • the browser’s user-agent text;
  • a label for the network an open came from, for some opens;
  • a random identifier stored in the visitor’s browser on your site, so a returning visitor is recognised.

We also hold your own account holders’ email addresses so they can sign in. For that we act as a controller, and the privacy page and our terms apply.

4. What we will do

  • Process the data only on your documented instructions: this agreement, your settings, and the calls you make through the app and the API.
  • Keep the data confidential, and make sure the people who can reach it are bound to confidentiality.
  • Apply the security measures in Annex 2.
  • Use only the sub-processors on the sub-processor page, and update that page before adding or replacing one so you can object.
  • Help you answer people’s requests: reading a person’s journey and events through the API, erasing a person with one call, and the public opt-out a visitor can use themselves.
  • Help you with impact assessments and with the supervisory authority, where it concerns this processing.
  • Tell you without undue delay after we become aware of a personal data breach affecting your data.
  • When the agreement ends, erase your data from the live database. You can delete your account yourself, at any time, from Settings, which erases it straight away. If you cannot (for example it manages client accounts), ask us and we do it ourselves, by hand, within 30 days of your request. You can read your data back through the API first, and you can erase individual people yourself at any time. Copies in backups age out as the privacy page describes.
  • Give you the information needed to show we comply, and allow audits as Article 28 requires, on reasonable notice.

5. What you must do

  • Have a lawful basis for every person you track, and give them the information GDPR requires.
  • Get consent on your own website before our script follows a visitor, because it stores an identifier in their browser (ePrivacy rules, Article 5(3)).
  • Give us lawful instructions only.

6. Where the data is

The service runs on servers in Frankfurt, Germany. We do not move tracking data outside the EU. If an off-site backup is switched on, the sub-processor page will name it and its region before it starts.

Annex 1. The processing

Subject matter, duration, nature, purpose, people and data are as in sections 2 and 3. The retention periods are on the privacy page.

Annex 2. Security measures

  • Traffic to the app and to tracking links is encrypted in transit.
  • Each account’s data is separate: every request is scoped to the account that made it.
  • API keys are stored only as a hash, with a short visible prefix.
  • An IP address is kept in one database column only (and in one index on it that is emptied in step with it), and our web-server logs drop the address before a line is written.
  • Retention is enforced by automatic sweeps, not left to a policy document.
  • Erasure removes a person from the live database, and a test enforces that every table holding a person’s data is covered.
  • The database is copied every night on the server. An encrypted off-site copy is built but not yet running.