Data Processing Agreement

updated 22 Nov 2026

1. What this agreement is

This is the data processing agreement under Article 28 of the GDPR between you, the customer, as controller, and Tim Clifford, trading as ReachKit, Schwedter Straße 235, 10435 Berlin, Germany (“ReachKit”, “we”, “us”), as processor. It covers the pages we host for you on your own domain, where we process the personal data of the people who visit them on your behalf. It forms part of our Terms and applies for as long as you publish hosted pages; there is nothing to sign. For everything else ReachKit does with personal data we are the controller, and our privacy notice applies.

2. The standard clauses

You and we agree to the European Commission’s standard contractual clauses between controllers and processors, set out in the Annex to Commission Implementing Decision (EU) 2021/915 of 4 June 2021, which are incorporated into this agreement by reference. Where they and our Terms disagree, the clauses prevail. In them:

  • Clause 5, the optional docking clause, is not used.
  • Under Clause 7.7, Option 2 applies: you give us general written authorisation to engage the sub-processors listed below, and we tell you of any intended addition or replacement at least 30 days in advance, so that you can object.
  • Clause 7.8 on international transfers is met by the safeguard named for each sub-processor below.
  • Their four annexes are the four sections that follow.

3. The parties (Annex I)

  • Controller: the customer who publishes hosted pages, as named in their ReachKit account, reached at the email address they sign in with.
  • Processor: Tim Clifford, trading as ReachKit, Schwedter Straße 235, 10435 Berlin, Germany, reached at hello@reachkit.app.

4. The processing (Annex II)

  • Subject matter and nature. Serving the pages ReachKit writes and publishes for you on a subdomain of your own domain, and counting how they are read.
  • Data subjects. The people, and the programs, that open your hosted pages.
  • Personal data. The visitor’s network address and the request headers their browser sends, such as browser details and the page they came from, processed while the page is served. We reduce the browser details, the referring page and any utm_source to a kind of read and its source: Google search, the named AI app, another site or opened directly. A click from a page to the business’s own site, booking, quote or phone link, which the visitor’s browser reports as it follows the link, is reduced to the kind of link. We keep only the count, the source or link, and the day, by page. The daily counts are kept for four weeks, and each page’s weekly counts for 13 months, with nothing that identifies a visitor. When a request claims to come from Google’s crawler, or from an AI service that publishes the addresses it uses, such as OpenAI’s or Perplexity’s, its network address is checked against that company’s published list and then discarded. A request that fails the check is not counted.
  • Special categories. None.
  • Purpose. To deliver your pages to their readers, to protect the service from abuse, and to show you how often AI answers and search engines read your pages, how people arrive at them, and how often they click through to you.
  • Duration. For as long as you publish hosted pages. When your access ends, hosted pages stay served for 30 days and then stop; when you delete your account they stop at once, and the counts are removed with the rest of its data within 30 days.
  • Frequency. Continuous, on every request to a hosted page.

5. Security measures (Annex III)

  • Hosted pages are served only over encrypted connections.
  • Stored data is encrypted at rest by our database and hosting providers.
  • A visit leaves nothing behind but a count: we store no visitor’s network address or headers, and we place nothing on their device.
  • Database access rules refuse every read and write by default; a customer’s data is reachable only from their own signed-in account and from our own servers.
  • Keys and credentials are held only as encrypted settings of our hosting provider, never in code and never in logs.
  • Access to the systems that run the service is limited to the operator.
  • We tell you of a personal data breach that affects your visitors without undue delay, and help you answer a visitor who exercises their rights. Since we keep nothing that identifies a visitor, there is usually nothing to hand over or delete.

6. Sub-processors (Annex IV)

This list is current as of the date above. Each sub-processor touches hosted-page traffic or what it leaves behind, and receives only what its job needs.

Who they are, and the one job each does for us
VercelServes every hosted page, and so receives each visitor’s request, network address and headers. United States; EU-U.S. Data Privacy Framework.
SupabaseHolds the pages and their daily and weekly counts; it receives no visitor’s address or headers. United States; Standard Contractual Clauses.
CloudflareRuns the name servers for the address your subdomain points at; it answers the lookup and does not receive the page request. United States; EU-U.S. Data Privacy Framework.

Before we add or replace a sub-processor, we change this list and its date, and email every customer with hosted pages at the address they sign in with, at least 30 days before the change takes effect. To object, write to hello@reachkit.app within that time. If we cannot meet your objection, you may unpublish your hosted pages or end your subscription before the change applies to you.

7. Contact

Questions about this agreement, and objections, go to hello@reachkit.app or to the postal address in our Imprint.