1. Information We Collect
We collect information that you provide directly to us, including:
- Account information (your email address, and how you sign in: with Google, or with an email and password)
- Payment information (processed securely through Stripe)
- Reading metadata — the reading type, the category you chose, and which cards were drawn
- Usage data and interaction with our services
Your question is not part of that by default. It is sent to OpenAI to generate the interpretation and then discarded — we do not write it to our database, and neither the question nor the interpretation is kept after the reading is returned to you.
Two kinds of reading
A reading you just take is ephemeral. The question and the interpretation exist for as long as the request takes and are then gone. We keep no copy, and there is nothing for us to hand over, lose, or be compelled to produce. If you close the tab, the reading is gone for us too.
A reading you save is stored on purpose. Pressing “Keep this reading” after a reading sends it back to us to be written to your account: the question, the interpretation, the cards, the spread, the category and the date. It happens only when you press that button, never as a side effect of reading.
Saved readings are visible only to you, at My Readings, where each one can be deleted individually. They are deleted along with everything else if you delete your account. We do not expire them on our own — a reading you asked us to keep is kept until you say otherwise.
1b. The daily card email list
If you join the waitlist on Card of the Day, we store your email address and nothing else — no name, no IP address, and no link to an account or to any reading. Tarotez sends no email at all today, so nothing is being delivered; the list exists so we can write once if daily delivery is ever built. Joining hands you a removal link you can use immediately, without an account.
1c. Feedback on a reading
After a reading we ask whether it resonated, and you can add a sentence saying why. Both are optional, and both are stored without anything that identifies you: no account, no email, no IP address, no question and no reading. There is nothing on that record to join back to you, which is also why answering does not require signing in.
What you write is kept as you wrote it, so please leave personal details out — we ask for this next to the box as well. The field is capped at a couple of sentences on purpose: it is there for a reason, not for the story behind your question.
1d. The morning reminder
If you switch on the reminder at Card of the Day, your browser creates a subscription and we store four things: the address your browser gives us to reach it, the two keys that let us encrypt a message only it can open, which language you were reading in, and how far your clock is from UTC — that last one only so the reminder arrives in your morning rather than ours.
That address identifies a browser, and we say so rather than pretending otherwise. It is not linked to an account, an email address or an IP address, and it is not joined to your readings. The notification itself contains an invitation and nothing else — not the card, which your device works out on its own when you open the page.
Turning it off in the same place removes the subscription from our side and cancels it in your browser. Blocking notifications for this site does the same thing from your end, and a subscription your browser has abandoned is deleted the first time we try to use it.
1e. What brought you here
When you create an account, we keep a note of how that visit started: the campaign
labels in the link you followed, if it had any (the utm_ tags we put in our
own ads and posts), the first page you opened on this site, and one word for where you
came from — search, social, another site, or directly. It tells us which of our ads
actually brought someone, so we can stop paying for the ones that did not.
It is not tracking across the web. No cookie, no advertising pixel, no IP address and no address of the page that sent you — only the labels we wrote ourselves and a page of our own. It is stored once, when the account is created, never updated, and deleted with your account.
1f. The server's own log
Like any web server, ours writes a line for each request it answers: the address requested, the answer it gave, how long it took, the IP address it came from, the kind of client (a browser, a search engine, the app) and — if you are signed in — the internal ID of your account, never your email. It is how errors get found and how we see which parts of the site are used.
It never contains what you wrote. No question, no interpretation, no note. The log rewrites itself as it fills, which at today's traffic means about a month.
1g. Counting the steps of a reading
The site counts, per day, how many times each step of a reading happens — a question started, cards chosen, a reading revealed, an account created — so we can see where people stop. Each count is the step, one value from a short fixed list (for example the kind of spread), the page language and the day.
A count cannot be traced to you. It carries no cookie, IP address, user identifier or time of day, and nothing that links one step to the next — and never your question, your cards or anything you write.
1h. Cookies, Google Analytics and Google Ads
Signing in sets two cookies that keep you signed in. They are needed for the account to work and are not used for anything else.
If you accept it in the cookie banner, we also use Google Analytics to
see how the site is used: which pages are visited, where visits come from and which steps
of a reading are reached. Google sets its own cookies (_ga, _ga_*,
kept up to two years) and receives the address of each page without its query string,
your browser and device, an approximate location derived from your IP address, and the
same steps described in 1g. Nothing of Google's loads until you accept. Google signals
are switched off.
The same consent covers Google Ads conversion measurement. If you arrived
from one of our ads, Google stores that click (_gcl_* cookies, kept up to 90
days) so that, if you then create an account, the ad can be credited with it. It is used to
know which ads are worth paying for — never to show you ads, build remarketing lists or
personalise advertising, which are switched off. That credit is a single conversion. The
page sends it once, only after the server confirms the account is new, and only if you
accepted cookies. Signing in again does not send it, and reloading the page does not send
it again.
It never receives what you wrote. No question, no interpretation, no note, no card names and no email. You can change your choice at any time from Cookie settings at the bottom of every page; rejecting removes Google's cookies.
1i. Payments and the free trial
When Stripe tells us a subscription payment went through, we keep a record of it: Stripe's invoice and subscription identifiers, the amount and currency actually charged, the plan and the date. We use it to know whether you have had the free trial (one per account), and to tell a first purchase apart from a renewal. A refund is recorded the same way.
Measured only with your consent. If you accepted cookies (1h), your browser tells Google Analytics, and Google Ads when you came from an ad, that a purchase happened, with the invoice identifier, the amount, the currency and the plan. Never your card, your name, your email, your question or your readings. Without consent nothing is sent.
2. Third-Party Subprocessors
Tarotez shares the minimum data needed to operate the service with the following subprocessors:
- Google Firebase (authentication): verifies your sign-in, with Google or with an email and password. Receives your email address and, if you use Google, your Google user ID. A password is stored by Firebase only as a salted hash and never reaches Tarotez. Firebase also sends the email that confirms your address and the one that resets your password.
- Stripe (payments): processes your subscription payment. Receives your email address and payment details; we never see or store your card number.
- OpenAI (tarot interpretations): receives your question and the cards drawn to generate an interpretation. We do not send your email, user ID, or any other identifying information along with the request.
- Sentry (error reports, servers in the EU): when the site or the server fails, receives a technical report of the error — the page path, your browser and the code that broke. Never the question, the interpretation, your email, cookies or IP address.
- Google Analytics and Google Ads (usage statistics and ad measurement, only if you accept the cookies): see 1h. Google may process this data outside the EU under the EU–US Data Privacy Framework.
We do not sell data or share it for advertising.
3. How We Use Your Information
We use the collected information for:
- Providing and improving our tarot reading services
- Processing payments and managing your account
- Communicating with you about our services
- Analyzing usage patterns to improve user experience
4. Data Security
We implement appropriate security measures to protect your personal information. This includes encryption, secure servers, and regular security audits. However, no method of transmission over the Internet is 100% secure.
5. Your Rights
You have the right to:
- Access your personal data (email support@tarotez.com for an export)
- Correct inaccurate data
- Delete your account and all associated data from your account page (Delete account button)
- Object to data processing
6. Contact Us
If you have any questions about this Privacy Policy, please contact us at support@tarotez.com.