Privacy Policy

The controller of your personal data — the party that decides why and how it is processed — is:

Trader

Petr Svoboda

Business ID

30052335

Address

Antala Staška 420/23140 00 Praha 4 - KrčCzech Republic

Contact

Support form

Register

Registered in the Czech Trade Licensing Register (Živnostenský rejstřík), kept by Úřad městské části Praha 4 (the Prague 4 trade licensing office).

We are a small operation and have not appointed a Data Protection Officer, which the law does not require of us. Privacy questions go straight to the address above and are answered by a person.

Information you give us

• Account details — your name, email address and a securely hashed password. We never store your password in readable form. If you add a profile photo, we store that too.
• Sign-in with Google or Microsoft — if you choose to sign in with one of them instead of a password, that company tells us your name, email address, profile photo and whether the address is confirmed, and we store the sign-in tokens it issues. With Google, the photo remains a link to Google’s servers, so your browser loads it from Google whenever it is shown.
• Your exam goal — which exam you are preparing for, the exam date if you enter one, and how thoroughly you want to prepare.
• Your answers — the responses you enter in exercises, including the text of Schreiben tasks and the text of Sprechen answers. Where you use the microphone, your speech is converted to text by the speech-recognition feature built into your browser (the Web Speech API, set to German). Depending on the browser, that conversion may run on your device or on servers operated by the browser’s maker (for example Google for Chrome, Microsoft for Edge), under that company’s own privacy policy — it is a feature of your browser, not a service we operate, and you can always type instead of speaking. We never receive, record or store audio of your voice; only the resulting text reaches us.
• Messages you send us — what you write through the support form (your name, email address, the topic, subject and message, and the page you were on) and any replies in that conversation, and what a school sends through the enquiry form on our schools page (school name, contact name, email address, optionally a phone number, the number of students, the exam, the timeline and a message).
• Exam-day reports — if you write one; see Exam-day reports.

Information created as you use the service

• Attempts and results — which exercise, Modellprüfung or level test you took, when, how long it took, your answers, your score, and the feedback produced for you.
• Felix’s free preview — if you use the one free marking per module, the full marking is kept on our server together with a fingerprint (a hash) of the marked text, so that the same answer can be unlocked later without being marked twice.
• A log of AI markings — the kind and time of each marking, to apply the fair-use limits.
• Activity record — for signed-in accounts, the pages you open and the links you click on our site; see Usage statistics.
• Where you came from — if you create an account, where the visit that led to it began, such as a search engine, another website or one of our ads; see Usage statistics.
• Technical session data — your IP address and browser user-agent, stored with your login session for security purposes. Our servers also use your IP address, in memory only, to limit how many requests can arrive in a short time, and write it to the server log when a request looks like an attack.

Information we receive from others

• Google or Microsoft, if you sign in with them (see above).
• Your school, if it has arranged access with us and asks us to set up accounts: the name and school email address of the people concerned.
• Stripe: whether and when your payment settled, the amount and currency, and Stripe’s references for the transaction.

Payment information

When you buy access, checkout happens on Stripe’s pages and your card details go directly to Stripe, which acts as the merchant of record — the legal seller of the payment transaction. We never see or store your card number. We pass Stripe your email address and your account reference so that the payment can be matched to your account. Stripe issues your receipt and invoice, and for payment data it is an independent controller under its own privacy policy. What we keep is the record needed to run your access: which plan you bought, when, the amount paid, the Stripe transaction references, and the date your access ends.

What we do not collect

We do not ask for and do not want special category data — health, beliefs, ethnicity or anything similar. Please do not include such information in Schreiben or Sprechen answers if you would rather it were not processed; the topics never require it.

Do you have to give us this data?

No law requires you to. The free exercises work without an account and without giving us anything. An account needs a name and an email address, buying access needs the payment details Stripe asks for, and Felix can only mark an answer he receives. Without that data, those parts of the service cannot work.

• To provide the service — running your account, saving your attempts, showing your history, progress and exam goal, marking your answers, the free preview of Felix’s marking, and the AI partner in the Sprechen Diskussion. Basis: performance of our contract with you (Art. 6(1)(b) GDPR).
• To grant paid access and keep records of purchases — recording which plan you bought and until when it runs, and keeping the accounting records the law requires. Payment itself is taken by Stripe as merchant of record. Basis: performance of the contract, and compliance with a legal obligation for the accounting and tax records (Art. 6(1)(b) and (c)).
• To keep accounts and the service secure — the session data and request limits described above, limits on how many emails one address can receive, and detecting abuse, account sharing and fraud. Basis: our legitimate interest in protecting the service and our users (Art. 6(1)(f)).
• To answer your messages and enquiries, including complaints and enquiries from schools. Basis: performance of the contract or steps before one, and our legitimate interest in answering whoever writes to us (Art. 6(1)(b) and (f)).
• To send service emails — confirming your email address, resetting your password, confirming what a purchase includes and when it ends, replies to your support requests, and, where the law requires it, notice of material changes to the service or these policies. (Your receipt and invoice come from Stripe, not from us.) Basis: performance of the contract and legal obligation (Art. 6(1)(b) and (c)). These are not marketing. We send no newsletters and no reminders about purchases you did not finish.
• To understand how the site is used — the usage statistics and the activity record described in Usage statistics. Basis: our legitimate interest in knowing where the site works and where people get stuck (Art. 6(1)(f)).
• To learn which channels bring registrations and purchases — where an account came from and the anonymous count of tagged arrivals, described in Usage statistics, so that we know which of our own ads are worth paying for. Basis: our legitimate interest in deciding where to spend on advertising (Art. 6(1)(f)).
• To improve the material — looking at which questions are answered wrongly far more often than expected, so we can find and fix bad items. For this we look at answers across all users, not at individuals. Basis: our legitimate interest in the quality of what we sell (Art. 6(1)(f)).
• To publish your exam-day report on our homepage, if you ask us to. Basis: your consent (Art. 6(1)(a)).

We do not sell your personal data, and we do not use it for advertising. Measuring which of our own ads lead to sign-ups shows you no ads and involves no ad network.

Felix, our AI examiner, marks Schreiben and Sprechen answers with an AI model operated by Anthropic PBC in the United States. The same provider plays the partner in the live Sprechen Diskussion. This is the single most important disclosure on this page, so here is exactly what happens:

• What is sent: the text of your answer, the task it responds to, and the language of the site you are using, so that the feedback comes back in it. In the live Diskussion, the conversation so far — your turns and the partner’s — is sent with each turn.
• What is not sent: your name, your email address, your account identifier, your payment details, or any audio.
• How it is used: only to produce your feedback and marks, or the partner’s next line, which are returned to us and shown to you.
• Training: under our agreement with the provider, your submissions are not used to train their models.

The partner’s voice. In the live Diskussion the partner’s lines are read aloud by a Microsoft text-to-speech voice. For this, our server sends Microsoft the text of the partner’s line and nothing else — not your voice, your name, your account or your IP address. The line is written by the AI, but it may take up an argument you made. We use Microsoft’s publicly available read-aloud service for this and have no separate data processing agreement with Microsoft for it. If the service cannot be reached, your browser reads the line itself.

No automated decisions. Felix’s marks are a practice estimate, not an exam result. They never decide your access, your price or anything else with a legal or similarly significant effect on you, so they are not an automated decision within the meaning of Art. 22 GDPR. The same applies to the automatic check of exam-day reports described below.

If you would rather nothing you write were sent to an AI provider, you can still use the whole of the Lesen and Hören material, which is marked against an answer key on our own systems — simply do not request Felix’s marking and do not start the live Diskussion.

We measure how the site is used, and what brings people to it, in five ways:

• Vercel Web Analytics and Speed Insights count page views and measure how fast pages load: the page address, the referring website, your browser, operating system and type of device, and the country derived from your IP address. According to Vercel, neither sets cookies. They also count a few named events — an account created, a checkout started, expired or paid (with the plan and amount), an exercise or level test completed — without your name, email address or account. The events carry where you arrived from (its source and medium, see below); the events for a new account and a completed payment also carry the campaign name.
• Where you arrived from. When you arrive, the page keeps in its memory where your visit began: the campaign tags of the link you followed (utm_source, utm_medium, utm_campaign, utm_term, utm_content) and whether it was a click on a Google ad — only that it was one: the click identifier Google adds to the link (gclid) is discarded and never stored or sent anywhere. Without tags, it is the domain of the referring website (search engines are recorded as “organic search”), otherwise “direct”; plus the first page you opened and the time. Nothing is written to your device — no cookie, no local or session storage — and the entry is lost when you reload or close the page.
• If you create an account. If you sign up during that visit — with the email form or through Google or Microsoft — the entry is sent with the registration and stored with your account: source, medium, campaign, keyword (utm_term), ad content (utm_content), whether it was a Google Ads click, the referring domain, the first page and the time you arrived. It tells us which channels, and which of our own ads, bring registrations and purchases, so that we can decide where to spend on advertising. It is not shared with Google or any ad network, not used to show you ads and not used for profiling; only we read it. It is kept as long as your account exists, deleted with it, and included in the data export on the Access & account page.
• An anonymous count of tagged arrivals. When someone arrives through a link with campaign tags or a Google Ads click, our server adds one to a counter per day and campaign (source, medium, campaign name and whether it was an ad click). No account, IP address or other identifier is stored with the count; the IP address is used only in memory, to limit requests, as described above.
• An activity record for signed-in accounts. While you are signed in, we store with your account the pages you open on our site and the links you click on them (the address and the link text), with the time, and the moments a Stripe checkout was opened, expired or paid. Never what you type, never your answers, nothing inside an exercise. Only we read it, to see where people get stuck. Entries older than 90 days are deleted; the deletion runs as new entries arrive for your account, so an account that is no longer used can keep older entries until it is deleted.

Basis: our legitimate interest in knowing how the site is used and which channels bring registrations and purchases (Art. 6(1)(f)). You may object to this at any time (Art. 21 GDPR) through our support form; if you object to the record of where your account came from, we delete it. What is stored on your device for these purposes is listed in our Cookie Policy.

School access. When a school arranges access with us, anyone who confirms an email address on the school’s domain receives that access automatically. For the arrangement itself we store the domain and its terms, not a list of pupils. If the school asks us to set up accounts or teacher roles, we do so from the names and email addresses it gives us and email the people concerned their sign-in details.

Joining a class. Entering a class code sends the teacher a request with your name and email address; nothing is visible until the teacher approves it. From then on the teacher sees your results at the class’s exam level (including Schreiben and Sprechen with their feedback), your flashcard progress, what you post in the class (messages, files, poll votes, reactions) and, for optional clubs, whether you said you are coming to the next lesson. Other members of the class see what you post there. A head of school sees counts and averages across the school’s classes, never a pupil’s name or individual result. Our support staff can look at a class read-only to answer the school’s questions. When you leave a class, the teacher stops seeing your work at once.

Teachers create classes, homework and flashcard decks and upload files, all stored in our database. A teacher can have a flashcard deck drafted by AI from a text they paste in; that text is then sent to Anthropic as described above.

Basis: performance of the contract for the class features you choose to use (Art. 6(1)(b)). Where a school arranges access for its pupils, the school decides, as controller, on what legal basis its pupils use the service, and its own privacy notice applies alongside this one.

Once your exam date or your access period is behind you, an account that bought access can write a short report — whether you passed, and how it went in your own words — and, separately, private feedback for us. If you leave “show it on the homepage” switched on when you send it, the report appears on our homepage with your name, your profile photo, the exam and the result as you stated it. The private feedback is never published.

Before a report appears, an AI model (Anthropic) checks the text for things such as insults, contact details or advertising. The check receives the text, the exam and whether you say you passed — not your name. If the check holds a report back, a person reads it before anything is published.

We also publish reports that people sent us in other ways, such as by email — only with their consent, under the name they chose and with a photo only if they sent one.

Basis: your consent for the publication (Art. 6(1)(a)); our legitimate interest in the check and in reading your private feedback (Art. 6(1)(f)). You can withdraw your consent at any time by switching publication off or deleting the report in your account; this does not affect the lawfulness of the publication before that (Art. 7(3) GDPR).

We keep this list short on purpose. This is everyone who receives personal data from us when you use the service:

Vercel, Neon, Anthropic and Resend process data only on our instructions, under written data processing agreements, and may not use it for their own purposes. Stripe is different: as merchant of record it is the legal seller of your purchase, and it processes your payment data as an independent controller under its own privacy policy — not on our instructions. Microsoft receives only the partner’s lines for the read-aloud voice, as explained in AI marking and speech, without a processing agreement.

Some companies receive data from you directly rather than from us, under their own privacy policies: Google or Microsoft when you choose to sign in with them, and the maker of your browser when its speech recognition runs on their servers.

We may also disclose data to our accountant, and to a public authority or court where the law requires it. If the business is ever sold or transferred, your data would pass to the buyer on the same terms, and we would tell you first.

Some of the companies above process data in the United States, as marked in the table. Those transfers are protected by the European Commission’s Standard Contractual Clauses, or by the provider’s certification under the EU–US Data Privacy Framework, together with technical measures such as encryption in transit.

You can ask us for a copy of the safeguards that apply to a particular transfer through our support form.

• Account details — while your account exists. Deleting your account — on the Access & account page, or by writing to us — removes them immediately and permanently, together with everything stored with the account: attempts, feedback, Felix’s previews, the log of AI markings, your exam goal, the record of where you came from, your activity record, class memberships and posts, your exam-day report and your support conversations.
• Attempts, answers and feedback — while your account exists, because your study history is the point of having one. You can have individual attempts deleted at any time by writing to us.
• Activity record — 90 days, as described in Usage statistics.
• Where you came from — while your account exists; deleted with it, or sooner if you object.
• Login sessions and the IP address stored with them — until the session expires (about seven days after you last used the site) or you sign out. Server logs are kept by our hosting provider only for a short time.
• Purchase and accounting records — the record of your access in our database is deleted with your account. Accounting documents that the law requires us to keep are kept for 10 years from the end of the accounting period, as Czech tax law requires; we cannot delete these on request. Stripe keeps its own records of the transaction under its own obligations.
• Support messages — up to 3 years, so we have the context of an earlier problem if it recurs; sooner if you delete your account.
• School enquiries — up to 3 years after the last contact; if an arrangement with the school follows, for as long as it runs.
• Exam-day reports — until you delete the report or your account.

Under the GDPR you have the right to:

• access your personal data and receive a copy of it (Art. 15);
• correct anything inaccurate (Art. 16) — you can change your name and photo yourself on the Access & account page; for anything else, write to us and we will fix it;
• delete your data (“right to be forgotten”, Art. 17) — you can delete your whole account yourself on the Access & account page — except where we must keep it, such as accounting records;
• restrict processing (Art. 18);
• portability (Art. 20) — download your profile, exam goal, where you came from, purchases, the log of AI markings, your exam-day report and all your attempts with their answers and feedback as JSON from the Access & account page; anything else we hold we send you on request, or to another provider if you ask;
• object (Art. 21) to processing we base on legitimate interests (see below);
• withdraw consent at any time, where processing is based on consent, without affecting what was done before (Art. 7(3)).

Your right to object under Art. 21 GDPR: where we process your data on the basis of legitimate interests (Art. 6(1)(f)) — for example the security measures, the usage statistics or the check of exam-day reports — you may object at any time on grounds relating to your particular situation. We then stop, unless we can demonstrate compelling legitimate grounds that override your interests, or the processing is needed to establish, exercise or defend legal claims.

To use any of these, contact us through our support form, preferably from the email address on your account. We will respond within one month, free of charge. We may ask you to confirm your identity before we act, so that nobody else can request your data.

If you think we have handled your data badly, please tell us first — but you have every right to go straight to a supervisory authority (Art. 77). In the Czech Republic that is the Office for Personal Data Protection (Úřad pro ochranu osobních údajů), Pplk. Sochora 27, 170 00 Prague 7, www.uoou.cz. You may also complain to the authority of the EU country where you live or work — in Germany, for example, the data protection authority of your federal state.

DeutschCertPrep is not intended for children under 15, and accounts may not be created by them. Purchases must be made by someone aged 18 or over.

If you are a parent or guardian and believe a child under 15 has given us personal data, contact us through our support form and we will delete the account and its data promptly.

Traffic to the site is encrypted with HTTPS. Passwords are stored only as salted hashes. Access to the production database is limited to the operator of the service, and the administration area requires two-factor authentication. Our hosting and database providers hold recognised security certifications and keep the data encrypted at rest.

No system is perfectly secure. If a breach ever occurs that is likely to result in a high risk to you, we will tell you without undue delay and report it to the supervisory authority within 72 hours, as the law requires.

We use cookies to keep you signed in and to remember your language, and your browser’s storage to keep your unfinished work on your own device. There are no advertising cookies and no third-party tracking across other websites. The detail, including what the usage statistics store, is in our Cookie Policy.

We will update this page when what we do changes — for example if we add a new provider. The date at the top always shows the current version. If a change materially affects how we use your data, we will tell you by email before it takes effect.