Privacy policy

Effective 11 October 2026 · applies to aipaylist.com

Short version: AIPayList shows no ads on this site, though we do buy Google search ads that point to it. We use page analytics to understand how people use the directory, including Google Analytics cookies outside the EEA, the UK and Switzerland. Apply and Join clicks are measured in Google Analytics, with those cookies where it uses them, and in PostHog without device storage. Our separate server-side click counts contain no user identifier, except the ad click identifier described in section 6 when you arrived from one of our ads and are not in the EEA, the UK or Switzerland. If you subscribe to alerts, send your CV for coaching, or write to us, we also process the email address, the CV and the messages you provide. If you rate a platform, we keep your answers and a scrambled key made from your IP address, never the address itself.

What we process, and why

1. Server logs (hosting)

The site is served by Vercel Inc. Like any web host, Vercel processes technical request data (IP address, user-agent, requested URL) transiently to deliver pages and prevent abuse. We do not export or analyze these logs at the individual level. Legal basis: legitimate interest in operating a website (GDPR Art. 6(1)(f)).

2. Server-side click counts

When you click “Apply” or “Join”, our server records that a click happened: a timestamp, which job or platform was clicked, the destination URL, whether it carries our referral code, and which section of the site it came from. In this separate database record, no IP address and no user identifier is stored with it. The one exception is the ad click identifier in section 6, stored only when you arrived from one of our ads and are not in the EEA, the UK or Switzerland. We use these counts to understand which listings people find useful. Legal basis: legitimate interest (Art. 6(1)(f)).

3. Aggregate page analytics

We use Vercel Web Analytics, a cookieless measurement tool: it reports aggregate page views and visit counts without setting cookies, without cross-site tracking and without storing IP addresses in a form linked to you. Campaign visits (for example a link from X with UTM parameters) are also counted with PostHog, hosted in the EU: it records page views, including the UTM fields and any ad click identifier on the URL, through a same-origin proxy so the browser does not call a third-party analytics host. PostHog runs in cookieless mode: no cookies and no local storage. When a visit arrives with campaign parameters (utm_* or a common click id), PostHog creates a hashed person profile so first-touch campaign attribution works, and keeps those fields on it, the click id included. That profile holds no name or email.

We have stopped the PostHog session replay test. No new session recordings are collected. During the test, about one page load in five was eligible for recording. Recordings held the page as it was drawn, clicks, scrolls and pointer movement. They excluded request contents, the browser console and canvas content. Typed input was masked, including email addresses and CV text in form fields. Visible page text was not masked, so chat messages already displayed on a page could appear in a recording. Pages opened from private confirm, manage or unsubscribe links were excluded. The test set no cookie and wrote nothing to device storage. Stopping the test does not delete earlier recordings. Legal basis: legitimate interest (Art. 6(1)(f)).

Google Analytics measures page views, sessions and visits using first-party analytics cookies, including _ga and _ga_KTD8G5N2FQ. These cookies can recognise the same browser across visits. In the EEA, the UK and Switzerland, Google Analytics uses no Analytics cookies at all, and neither sets nor reads them. There it sends the same page and click events without a browser identifier, so one visit cannot be linked to the next. Google decides where you are from your IP address, and where it cannot tell, the cookies are used. Your browser loads Google's measurement script and sends measurement events to Google. The page address in those events can include campaign fields such as utm_source and, if you came from one of our ads, the ad click identifier described in section 6. Google Signals and ad personalisation are disabled, and advertising storage is denied. We use these reports to understand which pages and traffic sources bring useful visits.

Google Analytics also receives Apply and Join click events with the job or platform identifier and the section of the page where the link appeared. Where the Analytics cookies are used, these events are associated with its browser and session cookies, so we can compare advertising visits with clicks through to hiring platforms. We do not send the outbound referral code, your CV, or your email address in these events. A click does not tell us whether you registered, applied, or were hired on the other platform.

During our homepage sorting test, a first-party cookie named apl_sort_v1 keeps the same browser in either the top-pay or newest group for up to 28 days. Its value is only the group name, not a unique identifier. When the job list becomes visible, Google Analytics receives the group, selected order, page number and whether filters were used. The tab stores this context in sessionStorage until it closes. We use the context for up to 24 hours after the first list view by adding it to later Apply click events from the same tab. Outside the EEA, the UK and Switzerland, Google Analytics also links those clicks through its cookies. This helps us compare which order leads to useful clicks.

PostHog also receives named events when you click Apply or Join, submit a search or filter form, open the chat window, or successfully submit the email-alert signup form. These events use fixed labels such as the section of the page, and a job or platform id for Apply and Join. They do not include search words, chosen filter values, names, email addresses, CV content, chat messages or referral codes. The filter, chat-opening and alert-signup events also go to Google Analytics, with its analytics cookies where it uses them. Automatic capture of other buttons and forms stays off in PostHog. These interaction events do not record the page or start session replay. Saving and removing roles remains unmeasured. A successful signup submission does not tell us that you confirmed your address, just as an Apply click does not tell us that you applied or were hired.

The CV template, the CV prompt and the one-click CV waitlist send named events of their own: asking for or downloading the template, copying the prompt, opening its preview, opening ChatGPT, Claude or Gemini from it, opening or joining the waitlist, and the optional answer about how many jobs you apply to in a month. These go to PostHog and to Google Analytics, with fixed labels such as the part of the site and the AI tool chosen, and with the address of the page, so a copied prompt is counted against the job it was for. They never include your name, email address, CV or the text of the prompt.

Clicking the link to the AI Train Station newsletter sends one named event too, to PostHog and to Google Analytics, with a fixed label for where the link sits (the home page, a job page, the CV template or the Mercor guide) and the address of the page. The sign-up itself happens on the newsletter's own page at beehiiv, under beehiiv's privacy policy. This site never sees the email address you give there.

4. Email alerts (only if you subscribe)

If you sign up for job alerts we store your email address, your chosen filters (minimum rate, categories) and, if you add one, your CV text, and use them solely to send you the digests you asked for. Every email contains an unsubscribe link; unsubscribing deletes the subscription. We never sell, rent or share subscriber lists. Legal basis: consent (Art. 6(1)(a)), withdrawable at any time. Details in the alerts section below.

5. Telegram alerts (only if you start the bot)

If you use our Telegram alert bot, we store your Telegram chat ID and your chosen filters so the bot can message you. Sending /stop deactivates the subscription. Telegram's own privacy policy applies to the Telegram service itself. Legal basis: consent (Art. 6(1)(a)).

6. If you arrived from one of our Google ads

Google adds a click identifier to the link you followed. If you are not in the European Economic Area (EEA), the United Kingdom (UK) or Switzerland, we keep it in a first-party cookie named apl_ad_click for 30 days, and if you then click Apply or Join we store it with that click so we can tell Google Ads which ad led to it. A Google Sheet we keep reads a file from our server that lists only those identifiers, each with its click time and a fixed conversion name, and Google Ads imports the file from that sheet. No email address, phone number or IP address goes with them, hashed or not. We do not use enhanced conversions, remarketing or audience lists. Google Ads receives them as an independent controller (see Where data lives, below), and Google's use of this data is described at policies.google.com/technologies/partner-sites. Separately from the cookie, the address of the page you land on carries the click identifier, and like any page address it reaches Google Analytics and PostHog as described in section 3, wherever you are. Inside the EEA, the UK and Switzerland, and wherever we cannot tell your country, the cookie is never set, and one set before we added this check is deleted rather than read, because it would need your consent there and we do not ask for it. Here that area also covers places with a country code of their own: Åland, Svalbard, Gibraltar, Jersey, Guernsey, the Isle of Man and the EU's outermost regions, such as Réunion and Guadeloupe. If you did not come from an ad, no such cookie is set. Legal basis when you are not in the EEA, the UK or Switzerland: legitimate interest in measuring our own advertising (Art. 6(1)(f)).

7. Messages you send us (only if you write)

If you use the message button in the corner of the page, we store what you wrote, your email address and the page you were on, and we forward the message to the operator through Telegram. It is stored because a reply typed on a phone has to wait somewhere until your browser next asks for it, which is what lets an answer appear in the chat while you are still reading. If you have already left the page when the reply is written, we email it to the address you gave, delivered by Resend. If you answer that email, your answer is stored in the same conversation and shown to the operator, the same as a message typed on the site. We use your address only to reply, never to add you to any list. While a conversation is open your browser also keeps a random conversation number in sessionStorage: that is not a cookie, it identifies the conversation rather than you, and it disappears when you close the tab. Legal basis: consent (Art. 6(1)(a)). Ask us to delete a conversation at any time and we will.

8. Saved roles (only if you save one)

If you press the bookmark beside a role, your browser keeps that role's id, title, platform, the pay it showed and the day you saved it in its own local storage, under the name apl_saved_jobs_v1. The list holds at most 100 roles. It is never sent to us or to anyone else, it is not a cookie, and no analytics event is sent when you save or remove a role. The bookmarks, the saved count and the list on the saved roles page were also left out of the replay test described above. Removing a role, or clearing this site's data in your browser, deletes it. It is stored only because you asked for it. The list itself never reaches us. Opening a saved role is an ordinary page view, measured as described above.

9. Coaching (only if you send your CV)

If you send your CV on the coaching page, we store your email address, the CV file, any note you add, and the exact consent text you agreed to and when. The file is kept in private storage that only our server can open. During this pilot, Alex reviews coaching CVs manually. They are not sent to an AI service.

Our server emails Alex with your CV attached. The email is sent by Resend and reaches Alex's Gmail inbox through Cloudflare Email Routing, so a copy of your CV is kept there too. Alex decides whether to write to you, and any coaching and its price are agreed with you by email.

We delete your CV and the rest of the record 12 months after you send it, unless you are working with Alex by then. The copy in Alex's inbox is deleted on the same schedule. Legal basis: consent (Art. 6(1)(a)). You can withdraw it at any time by writing to contact@aipaylist.com, and we then delete your CV and everything read from it, including the inbox copy. Any CV text or AI notes saved before this manual pilot follow the same deletion schedule.

Tool downloads (optional)

Some tools, such as the CV template, ask for your first name and email address. We use those two to send you the download and an occasional email when we add a new tool or feature. We also store which tool you asked for, the page you were on, and whether the download link was followed, which we count to see how the tool is doing. A mail scanner can follow a link before you do, so that last one counts link fetches, not people. We keep the exact consent text you agreed to and when, as the record of your consent. We use all of it for nothing else. Legal basis: consent (Art. 6(1)(a)), which you withdraw at any time with the unsubscribe link in every email. Unsubscribing deletes the record. Email is delivered by Resend, as for job alerts below, and apart from Resend we never sell, rent or share this list.

The waitlist for the one-click tailored CV works the same way: a first name and an email address, kept to say when it is ready to test and, now and then, that a new tool or feature is out, with the page you joined from counted. Unsubscribing from either list deletes your name and email from both. The CV prompt sends no part of your CV, and none of the prompt, to AIPayList. Your CV goes only to the AI tool you paste the prompt into. Pressing the prompt's buttons is only counted, as section 3 describes.

Email job alerts (optional)

If you sign up for alerts at /alerts, we store your email address, the preferences you set (minimum rate, kinds of work, countries, expert field, digest timing, any title keywords), your CV text if you added one, the exact consent text you agreed to and when, and the timestamps of your confirmation and the last digest we sent. Legal basis: consent (Art. 6(1)(a)), which you withdraw at any time by unsubscribing.

If you choose to add a CV, we keep its text together with your subscription and use it for one purpose: building a skills profile (skills, fields, seniority, languages, countries you can work from, and a one-sentence summary with no personal identifiers) so your alerts match you better. The uploaded file itself is discarded after we read its text. The profile is built either by our operator reading the text, or by an automated extraction using Google's Gemini API, which receives the CV text and nothing else about you. Unsubscribing deletes your CV text along with the rest of your record, and you get the same matching without a CV by filling in the form fields instead.

Nothing is sent until you confirm your address (double opt-in). Every email carries an unsubscribe link, and unsubscribing deletes your record outright rather than marking it inactive. Unconfirmed signups are deleted after 7 days; subscribers we have not emailed for 12 months are deleted. Emails are delivered by Resend on our behalf (Resend sends through Amazon SES in the EU, Ireland region) and contain no tracking pixels. Mail you send to contact@aipaylist.com is routed to us by Cloudflare Email Routing, which also provides our DNS.

Worker ratings (optional)

If you rate a platform you have worked on, we store your five answers, the platform and the time. To keep one rating per connection per platform we also store a scrambled key computed from your connection's IP address and the platform with a secret only our server holds. The IP address itself is never stored, and the key cannot be turned back into it without our secret. A second rating from the same connection replaces the first, so people who share a connection (an office, a VPN, some mobile networks) share one rating. Each rating is also sent to the operator as a Telegram message with the platform and the five answers, and no key or address, so every rating is seen as it arrives. The platform's page shows how many ratings it has from the first one. The summary (the average and the shares) appears only once there are 10 ratings from the last 12 months, and each share only once it rests on 10 ratings, so no average or share on the page rests on fewer. A rating stops counting after 12 months, but nothing deletes it on a schedule, so it stays in our database until we remove it. Legal basis: legitimate interest in publishing honest worker ratings and keeping them free of repeat voting (Art. 6(1)(f)). To change your answers, rate again from the same connection. To have a rating removed, email us the platform, the day you last rated it and the five answers you gave, which is how we find it.

Where data lives

Our database (job listings, click counts, alert subscriptions, tool download requests, chat conversations, worker ratings, coaching applications and their CV files) is hosted with Supabase in Frankfurt, Germany (EU). The website is delivered by Vercel's edge network with primary serving from the EU (Frankfurt region). Cookieless pageview events used for campaign attribution are processed by PostHog in the EU. Page and session measurement also reaches Google Analytics, and Google processes it outside the EU under the EU-US Data Privacy Framework and standard contractual clauses. Outside the EEA, the UK and Switzerland, Google Analytics events include the browser identifiers described above. CV text used for alert profiles reaches Google's Gemini API on the same terms. These providers act as processors under data-processing agreements. The copy of a coaching email in Alex's inbox is held by Google's Gmail under Google's own terms. Google Ads is different. The ad click identifiers and click times in section 6 pass through a Google Sheet we keep, and Google Ads receives them as an independent controller under the Google Ads Controller-Controller Data Protection Terms, not as our processor.

What happens when you click through to a platform

Every listing links to the hiring platform's own site. From the moment you land there, that platform's privacy policy governs. Some outbound links are referral links (see the disclosure page): the platform may credit our account for the referral and may show us referral status and candidate details in its referral dashboard.

What we don't do

No ads on this site. No email addresses or phone numbers sent to Google Ads, hashed or not, no remarketing and no audience lists. Google Signals and ad personalisation are off and advertising storage is denied on the one Google tag we load. No fingerprinting. No sale of personal data, and no sharing of it beyond what this page describes. No automated decision-making with legal or similar effects. The only profiling is the skills profile built from a CV you choose to add to alerts. Coaching CVs are reviewed manually by Alex, who decides whether to write to you. We never ask for special-category data, such as health, religion or ethnic origin, so please leave it out of your CV. If a CV you send mentions any, it is stored and read only as part of the CV and deleted with it. The service is not directed at children under 16.

Your rights

Under the GDPR you can request access to, correction of, or deletion of your personal data, restriction of or objection to processing, and data portability. The personal data we hold about you is mostly what you gave us: an email or Telegram alert subscription, a tool download request, a coaching application, a worker rating or a conversation you started. A worker rating is kept under the scrambled key described in Worker ratings above, and we find it for a request by the platform, the day you last rated it and your answers. The data we hold also includes the ad click identifier in section 6 (on a click record only for visitors who are not in the EEA, the UK or Switzerland, and in our PostHog and Google Analytics data for every visitor from our ads). The practical route is to unsubscribe (instant deletion), or to email us for anything else, including deleting a coaching application. You also have the right to lodge a complaint with your local supervisory authority.

Contact

AIPayList is an independently operated project. Data controller contact: contact@aipaylist.com.

We'll update this policy if our practices change, with the effective date above revised accordingly.