Privacy Policy

Last updated: 26 October 2026

This policy explains what personal data TakeJournal keeps when you use the web site takejournal.com and the app at app.takejournal.com, why, for how long, and what rights you have. We keep only what the service needs, we never sell your data, and we do not use it for advertising.

1. Who is responsible

The controller of your personal data is IT System Development s.r.o., Roztylská 1860/1, 148 00 Praha 4 – Chodov, Czech Republic, company ID (IČO) 07553897, VAT ID CZ07553897, registered in the Commercial Register kept by the Municipal Court in Prague, Section C, Insert 302019 (“we”). You can reach us at support@takejournal.com about anything in this policy.

2. What we keep

• Your account: e-mail address, name, and your password, stored only as a salted hash. If you sign in with Google, also your Google account’s identifier, e-mail address and name.
• Google sign-in: when you sign in with Google, Google tells us your Google account’s identifier, e-mail address, whether that address is confirmed, and your name. It may also send your profile picture and language, which we do not keep. We use these data only for your TakeJournal account: to sign you in, to show your name and address in the app, and to send you the e-mails about your account described below. We do not share them with anyone and do not use them for anything else, and they are deleted with your account. TakeJournal’s use of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.
• Broker connections: the login you enter for your broker or its data provider (for example Rithmic): user name, password, system and gateway; or, where the platform issues one, your API key (for example TopstepX). The password or key is stored encrypted. We use the login only to read your data. TakeJournal never places, changes or cancels an order. Where the platform lets you sign in on its own page instead (for example tastytrade, cTrader or Tradovate), we never see your password: the platform gives TakeJournal an access token, which we store encrypted and use only to read; you can withdraw it at the platform or by deleting the connection.
• The NinjaTrader 8 add-on and the MetaTrader 4 and 5 programs: here you give us no password. You install TakeJournal’s add-on in NinjaTrader 8, or its program in MetaTrader 4 or 5; it carries a key of its own, of which we keep only a hash, and sends us your trading account’s executions or deals (the account, the instrument, the side, the quantity, the price, the time, the order and the fees) and, from MetaTrader, the account’s deposits and withdrawals and the contract sizes of the symbols traded, nothing else. A new key, or deleting the connection, stops the old one.
• Your trading data: what we read from your broker or what you import from files: your trading accounts (names, balances, risk limits), orders and fills, the trades built from them, results and fees, and the market prices around each trade used for charts and statistics.
• What you add: notes, tags, strategies, ratings, day and week notes, cash records such as payouts and fees, screenshots, and trades you enter by hand.
• The AI import: when you ask the AI to map a statement whose format TakeJournal does not know, we send Anthropic the file’s name, its column names and its first twenty rows, together with your time zone; for a PDF statement, the text of its pages. Before anything leaves our server, account numbers, other long identifiers and e-mail addresses are replaced by placeholders. A PDF statement’s header may still show your name and address as the broker prints them; they go with the text. The answer (which column means what, or the fills found in a PDF) is applied on our server to the rest of the file. We keep a record of each such request: when, for whom, how many tokens it used and what it cost, never the request’s content. Nothing is sent unless you choose the AI in the import dialog; without it, a statement is mapped by its column names on our server alone. The mapping it returns (which column means what, never a row of the file) may be offered to another user who imports a file with the same columns, so that their import needs no AI request. For a statement you mark in the import dialog as one from OKX, Bitget or Hyperliquid whose layout TakeJournal does not know yet, the same is sent without that choice, until the layout is learnt. Of such an import we keep its column names and the mapping you imported it with, linked to your account only until three users’ imports agree; then the mapping alone remains, so that the exchange’s statements import for everyone.
• The AI chat: when you ask the chat in the app a question, we send Anthropic the question, the page you ask from and its filters, and what the chat reads from your journal to answer it: figures, trades with their notes and tags, day and week notes, and your accounts’ kinds, firms and balances. Before anything leaves our server, account numbers and account names are replaced by placeholders, and the answer shows them again on our side. We keep your conversations so you can return to them, and the same record of each request as for the AI import. Nothing is sent unless you ask the chat a question.
• The guide chat on this web site: when you ask the chat on takejournal.com a question without an account, we send Anthropic the question and the address of the page you read, and the chat answers from our guide alone. We keep each conversation for 30 days after its last question: your questions as you wrote them (with anything you type into them, such as an e-mail address or a name), the address and the language of the page, and the answers, with when they were given. Our team reads them to improve the guide and the FAQ, and is told by e-mail of each new conversation, without its text. Nothing else about you is kept with them: the chat has no account behind it, so we know your name or e-mail address only if you write it. Of each answer we also keep how many tokens it used and what it cost, with a hash of a random identifier in a cookie (tj_visit) and a hash of your IP address, which count the questions per visitor and per network and keep a follow-up question in its conversation; both hashes are deleted after two days.
• The AI reviews: on a plan with AI reviews and with “Review every week by itself” switched on in your profile (it is off for a new account), after each week in which you closed trades we send Anthropic that week’s journal so that its AI can write the week’s review: the trades with their notes, tags, checklist answers and plans, your day and week notes, the week’s statistics and those of the 90 days before it, your prop accounts’ rule state and balances, on Fleet what Spotter told you during the week, and the previous review. A day’s review is sent only when you ask for it on the day’s page, with the same data for that day. Before anything leaves our server, account numbers and account names are replaced by placeholders. The review is kept in your journal like a note, and we keep the same record of each request as for the AI import.
• Spotter’s words: on a plan with Spotter and with “Worded by the AI” switched on in your profile (it is on unless you switch it off), each message Spotter says while TakeJournal is open goes to Anthropic so that its AI can word it in your language, with the message’s facts only: Spotter’s plain sentence, the trade’s symbol, size, times and result, the day’s trades and result so far, and the rule or the history the message quotes. No notes, no account names and no account numbers are sent. The worded message is kept with Spotter’s message in your journal, and we keep the same record of each request as for the AI import.
• Reports you send us: when you report a problem or an idea to us from the AI chat (it sends a report only when you ask it to and agree to what it will send) or from its form “Report a problem or an idea”, we keep what you wrote, the page you were on and its filters, the app’s version and, for a report from the chat, the conversation it came from, with your account. Only our team reads the reports.
• Pages you share: when you make a public link with Share (to a day, a week, a month or one trade), we keep a copy of what its page shows, frozen at that moment: the result in the unit you chose, for the accounts you chose together; if you list them, each trade’s time, symbol, direction and result; for one trade also its times, prices, points, R and the best and the worst price while it was open; and a name only if you type one. The page never shows an account’s number, firm or balance, your e-mail, sizes or notes. Anyone with the link can open it without signing in, and it asks search engines not to index it. We also keep when the link was made and revoked and how often it was opened, nothing about who opened it. A shared page may also be shown embedded in another web site where you place it.
• Files you import: we keep each file you import, compressed, so that you can download it again under Imports on its account’s page; it counts in your storage. It is deleted when you delete it there, when you delete the account it went into, when you start over, or with your account.
• The API and connected AI assistants: on a plan that includes them, you can create a key for TakeJournal’s API in your profile, or connect an AI assistant of another company (for example Claude or ChatGPT) to your journal through TakeJournal’s MCP server, by signing in on our page and allowing it. A key or an assistant can only read: your accounts, trades, fills and statistics, and, if you allow it, your notes, day and week notes, AI reviews and Spotter’s messages. It changes, deletes and sends nothing, and never reaches your broker. It reads until you revoke the key or disconnect the assistant in your profile, or until it expires. Of a key we keep only its hash, with its name, what it may read, when it was made, when it expires and when it was last used; of an assistant the same, with the name and the address it gives for itself, and only hashes of the access tokens we give it. We count the calls of each key and assistant to keep your plan’s limits. Whoever holds a key can read your journal through it, so keep it secret. What an assistant reads goes to the company that provides it, with what you ask it, under that company’s own terms and privacy policy (see section 5). An assistant reads your account numbers and account names only if you allow it when you connect it or make its key (the box is ticked to start with); otherwise they are replaced by placeholders, as in the AI chat. A key used for your own calls to the API reads them as the app shows them.
• People you invite: if you give someone (for example a mentor) access to your journal or a part of it, they see what you share with them until you end the access.
• A review by a person: only when you ask for it, someone of our team may read your journal to review it; they read what the review needs, and the review is kept in your journal.
• Your public profile: if you choose to publish a profile (for example verified results or a place on a leaderboard), it shows only what you choose, to the public or the community you choose, until you take it down. Nothing is published without your choice.
• Anonymous statistics: your trades and accounts may count in statistics over many traders that we may publish (for example the pass rates of the prop firms’ evaluations or the time to a first payout). They count without your name, your e-mail, account names or numbers, and are published only as totals from which no trader or account can be recognised. A switch in your profile, which comes with the first such statistics, leaves you out at any time; your data then counts in no new ones.
• Push notifications: when you turn on notifications on a device (Profile → Alerts → On this device), its browser gives us an address at its maker’s push service and the keys to encrypt for that device. We keep them, with the time the last notification was delivered, until you turn the notifications off on that device or sign out there, or the push service tells us the address is gone. Each notification (an alert, or on Fleet a message of Spotter’s) is encrypted so that only that device can read it, and carries no amounts.
• Technical data: when you use the app our server records the IP address, the browser’s identification, the address requested and the time. The web site takejournal.com keeps no such log. If we need it to find a fault, the web site may keep the same log too, for at most 14 days.
• Payments: when you buy a plan, your name, billing address and VAT ID if you give one, the plan and its period, your payments, invoices and refunds as Stripe reports them, and Stripe’s identifiers of you as a customer and of your subscription. Your card details go to Stripe only; we never see them. A pause of your subscription is kept with its start and its end, and we e-mail you when you set it and a week before it ends. When you cancel a subscription in the app and say why, we keep the reason you chose and your comment with the plan; our team reads them.
• Invitations: with a paid plan your Profile shows your own invitation link with a code we make for you. When someone makes an account through such a link and subscribes within 30 days, we keep whose link brought them and when, and the reward of their first payment: its amount, the payment it comes from, when it is due and when it was credited to the inviter’s balance at Stripe. The inviter sees, for each trader their link brought, when they came, their plan, whether it is a trial or paid, and the reward, never their name or e-mail address. The invited trader sees nothing about the inviter.
• Partner codes: some promotion codes belong to a partner of ours, a web site or a creator who recommends TakeJournal. When you subscribe with such a code, we keep which code it was and the payments made with it, so that we can pay the partner its commission for your first 12 paid months. The partner gets from us only monthly totals: how many customers used the code, what they paid and the commission; never your name, your e-mail address or anything from your journal.
• Partners: if you are our partner, your name or your company’s, how to contact you, your payment details (a bank account or a Wise account) and your invoices to us.
• What you agreed to: when you created your account and which versions of our terms and of this policy you agreed to; your choice about product news by e-mail and each change of it; and, for each purchase, your request at the checkout that the plan start at once.
• E-mails to you: e-mails about your account and your subscription: confirming a sign-up, resetting a password, an invitation someone sent you, a reminder before a yearly plan renews, a confirmation when your account is deleted. Product news only if you did not refuse it at sign-up and have not switched it off later on your Profile page; each such e-mail has a link that stops them. Product news includes up to three e-mails after a paid plan ends (7, 30 and 90 days later) about what your account keeps and about our plans; the one after 30 days follows the reason you gave when cancelling, if you gave one. Of each e-mail we send we keep to whom, which message, when, and whether it was delivered, bounced or marked as spam. When you unsubscribe, or when an e-mail to you bounces for good or is marked as spam, we keep a one-way hash of your address (not the address itself) on a list of addresses that get no product news; this list stays after your account is deleted.

3. Why, and on what legal basis

• To provide the service you signed up for (performance of a contract, Art. 6(1)(b) GDPR): your account, reading and showing your trading data, the statistics computed from it, the public pages you choose to share, your subscription and its payments, and e-mails about your account and your subscription.
• To run the referral programme (performance of a contract, Art. 6(1)(b) GDPR, as the terms describe it): whose link brought whom, the rewards, and showing an inviter the plan and the reward of the traders their link brought, which a trader accepts by subscribing through the link.
• To pay our partners their commission (our legitimate interest, Art. 6(1)(f) GDPR, in paying partners what we owe them): which partner code a subscription was paid with and the payments made with it.
• To work with our partners (performance of a contract with the partner, Art. 6(1)(b) GDPR): a partner’s name, contact and payment details; their invoices are accounting records (below).
• To keep accounting and tax records (a legal obligation, Art. 6(1)(c) GDPR): invoices and payments.
• To prove what you agreed to and what we sent (our legitimate interest, Art. 6(1)(f) GDPR, in showing that we keep the law): the record of what you agreed to and the log of e-mails.
• To send product news to our customers (our legitimate interest in telling customers about our own service, Art. 6(1)(f) GDPR, as Czech Act 480/2004 Coll., § 7(3) allows): only to users who did not refuse it; you can refuse it at any time, free of charge, with the link in every such e-mail or on your Profile page.
• To never send product news to an address that refused it (a legal obligation, Art. 6(1)(c) GDPR): the list of hashed addresses.
• To keep the service secure and working (our legitimate interest, Art. 6(1)(f) GDPR): server logs, backups, and limits on repeated sign-in attempts. To limit repeated requests from one network we count them by IP address, for a few minutes, in memory.
• To keep the Free plan fair (our legitimate interest, Art. 6(1)(f) GDPR, in a free plan that one person cannot multiply): once a day our server compares the trades of the last 30 days on each account with the Free plan or a trial with the trades of the other accounts – symbol, direction, the minute of entry and of exit, size and prices – and notes when two accounts’ e-mail addresses lead to the same mailbox (for example a Gmail address with dots or a +tag). When most of two accounts’ trades match, a person on our team looks at the match; nothing is decided automatically, and an account is suspended only after that look, with a message telling you why. No copy of your trades is made for it, and no IP address or device is recorded for it.
• To fix and improve TakeJournal from a report you send us (our legitimate interest, Art. 6(1)(f) GDPR, in acting on what you reported): the report and what goes with it.
• To answer a question you ask the guide chat on the web site (our legitimate interest, Art. 6(1)(f) GDPR, in answering what you asked): the question, and the hashes that limit how many questions a visitor and a network may ask.
• To improve the guide and the FAQ from what visitors ask the guide chat (our legitimate interest, Art. 6(1)(f) GDPR, in a guide that answers what visitors want to know): the conversations, for 30 days, read only by our team. You can object, or ask us to delete a conversation sooner, at support@takejournal.com; tell us what you asked and when.
• To provide what you choose to use (performance of a contract, Art. 6(1)(b) GDPR): the API and the assistants you connect, the people you invite, the files you keep, your public profile and a review you ask for.
• To compute and publish anonymous statistics over many traders (our legitimate interest, Art. 6(1)(f) GDPR, in showing traders verified figures about the firms and platforms they use): your trades counted without anything that identifies you; you can object at any time with the switch in your profile.

We do not profile you, we do not use analytics or tracking services, and we show no advertising. We count the visits of the web site and the app ourselves, without cookies (our legitimate interest, Art. 6(1)(f) GDPR, in knowing how many people use TakeJournal): from each page load our server forms a key out of the IP address, the browser’s identification and a value that changes every day and is never stored, and keeps only how many different keys and page loads each day had. Neither the key nor the address is stored, and the keys of different days cannot be linked. A visit counts only once the page has run in your browser: after it has loaded, the page sends our server its own address and nothing else, and the key is formed from that message the same way. When you arrive from another site or from a link, the page also tells our server where you came from: the ref or utm_source of the link you followed, else the name of the site that linked to it (for example betalist or google), else that you came directly. We count, per day, how many visitors and arrivals each such source had, and nothing more. If you then make an account, we keep with it the first source your browser brought, only its name, to know which places bring traders to TakeJournal; our team sees it only as counts per source.

The Free-plan check above is the only comparison between users’ data we make, and it decides nothing by itself.

4. Cookies and browser storage

The app sets two cookies, both needed for it to work:

• TAKEJOURNAL_SESSION keeps you signed in. It ends after 30 days without activity or when you sign out. Its name may carry the prefix __Host-, which lets only app.takejournal.com set it.
• XSRF-TOKEN protects your actions in the app against requests from other sites.

The app also remembers your display choices (theme, chart layers, filters) in your browser’s storage; they never leave your device. The web site takejournal.com sets one cookie, and only when you ask its guide chat a question: tj_visit, a random identifier that counts your questions, for a day. There are no third-party cookies, so we ask for no cookie consent. The web site and the app also keep, in your browser’s storage for 30 days, the name of the first source you came from (tl.source, for example betalist, with the time); the web site adds it to its links into the app, and it leaves your device only with those links and with a sign-up.

A promotion code or an invitation code that a link brought is kept in your browser’s storage for 30 days; it reaches us only with a checkout and is forgotten after it.

5. Who processes data for us

• OVHcloud runs our server and the storage of the screenshots and of the files you import (its object storage, encrypted) in Limburg, Germany (EU), and our domain names.
• Google (Google Ireland Limited) provides Google sign-in when you choose it and our own mailboxes (Google Workspace). Google may process data outside the EU; it does so under its data processing terms, which include the EU standard contractual clauses.
• Postmark (ActiveCampaign, LLC, USA) sends our e-mails to you and tells us whether each was delivered, bounced or marked as spam. It receives your e-mail address and the e-mail’s content, under its data processing addendum, which includes the EU standard contractual clauses.
• Stripe (Stripe Payments Europe, Limited, Ireland) processes payments, invoices, VAT and refunds when you buy a plan, and keeps your card details. Stripe processes some data as a controller of its own (for example to prevent fraud and to meet its legal duties as a payment provider), under its own privacy policy; data may go to Stripe, Inc. in the USA under the EU standard contractual clauses and the EU-U.S. Data Privacy Framework.
• Your broker and its technology provider (for example Rithmic) receive your login when we connect on your instruction, as they would from any trading software. Their own privacy policies apply to what they do.
• Anthropic (Anthropic, PBC, USA) answers the AI import’s, the AI chat’s (in the app and on the web site), the AI reviews’ and Spotter’s requests described above through its API. Anthropic does not use these requests to train its models and does not keep their content after answering, under its commercial terms and data processing addendum, which include the EU standard contractual clauses for the transfer. We send no account number: account numbers (and for the import e-mail addresses, for the chat and the reviews account names) are replaced by placeholders before the request is made.
• The company behind an AI assistant you connect (for example Anthropic for Claude, OpenAI for ChatGPT, Google for Gemini) is not our processor: you choose it, and it receives what the assistant reads from your journal, with what you ask it, under your own agreement with that company. Its terms and privacy policy decide where it processes the data (possibly outside the EU), how long it keeps them and whether it uses them to train its models. This is separate from Anthropic’s work for us above: an assistant you connect works for you. It receives nothing while no assistant is connected.
• The push service of your browser’s maker (Google for Chrome, Apple for Safari and for the app on an iPhone or iPad, Mozilla for Firefox, Microsoft for Edge) delivers the notifications to a device on which you turned them on. It receives each one encrypted for that device alone, so it cannot read it; it sees the address your browser gave us, when a notification is sent, its size and our server’s address. Your browser chooses the service, and its maker’s own privacy terms apply.

Daily backups of the server are also kept on our own storage in the Czech Republic. The database may also be copied continuously to a second server or storage of ours in the EU, so that a fault loses seconds of your journal rather than a day. If we move the server, the backups or such a copy to another provider, it is one in the EU, and we update this list before the move.

Some links to brokers or prop firms in TakeJournal may be affiliate links. If you follow one, the firm learns that you came from TakeJournal (an identifier of ours in the link), never your name, your e-mail or anything from your journal.

6. How long we keep data

• Your account and everything in it: as long as your account exists. You can delete it yourself at the foot of the Profile page, which deletes everything at once; when you ask us to delete your account instead, we delete it within 30 days; the copies in our backups disappear within 14 days after that.
• A screenshot or a broker connection you delete is gone at once; backup copies disappear within 14 days. Deleting a broker connection deletes its stored login.
• A device’s push address: until you turn the notifications off on that device or sign out there, or its push service reports it gone.
• Server logs: at most 6 months.
• Invoices and payment records: 10 years after the year of the payment, as Czech tax law requires (also after your account is deleted).
• The record of what you agreed to: as long as your account exists.
• The reason you gave when cancelling a subscription, and its comment: as long as your account exists.
• The log of e-mails we sent: 3 years; when your account is deleted, its entries stay with the address only, as proof of what was sent.
• The hashed addresses that get no product news: as long as we send product news at all.
• AI chat conversations: 12 months after their last question, unless you delete them sooner in the chat.
• Reports you send us: as long as your account exists; a report keeps no conversation you deleted.
• Pages you share: the copy as long as your account exists; Revoke in Profile → Data → Shared snapshots ends a link at once, and its page then shows nothing.
• Files you import, invitations and a public profile: until you delete or end them, or delete your account.
• API keys and connected assistants: until you revoke or disconnect them, they expire, or you delete your account. The count of their calls: for the day only.
• Anonymous statistics: the published totals hold nothing from which you can be recognised, and they stay; when you switch the statistics off or delete your account, your data counts in no new ones.
• The guide chat on the web site: each conversation 30 days after its last question; the hashes of the visitor and the network two days.
• Links in our e-mails expire: password reset after 1 hour, sign-up confirmation after 24 hours, invitations after 7 days.
• A match the Free-plan check found, with our decision about it: 12 months, or until one of the two accounts is deleted.
• Whose link brought whom, and the rewards: as long as both accounts exist; a credited reward also stays in Stripe’s record of the inviter’s balance.
• Which partner code a subscription was paid with, and its payments: as long as a commission can be due on them (the customer’s first 12 paid months, until the last month’s commission is paid); then only in the accounting records, for the 10 years above.
• A partner’s details: as long as the partnership lasts and a commission can be due; their invoices and our payments to them for the 10 years above.

7. Your rights

You can ask us for a copy of your data, to correct it, to delete it, to limit what we do with it, or to give it to you in a machine-readable form, and you can object to processing based on our legitimate interest. Write to support@takejournal.com; we answer within one month. You can also complain to a data protection authority, in the Czech Republic the Office for Personal Data Protection (uoou.gov.cz), or the authority where you live.

8. Security

All traffic is encrypted (HTTPS). Broker passwords are stored encrypted, the data lives on a server in the EU that only the service can read, and it is backed up daily.

9. Age

TakeJournal is meant for adults. Do not use it if you are under 18.

10. Changes to this policy

When we change this policy we publish the new version here with its date, and we tell registered users about important changes by e-mail before they apply.