← All policies

Privacy Policy

Version 1.12 · Effective date: 21 October 2026

Scinova Group LLP ("ReadNTag", "we", "us", "our") provides the ReadNTag reference-management service at readntag.com, including the web app; our browser extensions (such as for Chrome, Firefox, Safari, and Edge); any mobile or desktop applications; any add-ins or add-ons for third-party tools (such as Microsoft Office or Google Workspace); the conference and opportunity listings; and any other apps, integrations, features, and related tools we make available from time to time (together, the "Service"). Not all of these are available at any given time; this definition covers them where and when we offer them. This Privacy Policy explains what personal data we collect, why, how we protect it, and the rights you have under Singapore's Personal Data Protection Act 2012 ("PDPA") and, where applicable, the EU/UK GDPR.

1. Who we are

Scinova Group LLP (UEN T23LL1363K) is a Singapore Limited Liability Partnership at 60 Paya Lebar Road, #06-28 Paya Lebar Square, Singapore 409051. We are the data controller for personal data processed through the Service.

Our Data Protection Officer can be reached at dpo@readntag.com.

2. What we collect

  • Account data: your email address, display name, and a securely hashed password (we never store your password in plain text). If you choose to sign in with Google, Google shares your name, email address, and a profile identifier with us so that we can create or match your account. We never receive your Google password. Google's handling of your information is governed by Google's own privacy policy.
  • Your content: the references, PDFs and other files you upload or capture, together with the metadata and everything you create, generate, organise or edit using the Service's tools, including but not limited to topics, projects, highlights, notes, snaps, citations, matrices, screening decisions, reading lists, boards, planner entries, saved searches and feed views, and any similar content you produce through features we add over time.
  • Usage data: events needed to run, secure and improve the Service, for example logins, paper views, exports, storage used, and the other interaction events generated as you use the Service's features (such as the items you open, the actions you take, and the tools and features you use), together with session records (sign-in time and duration), the IP address and browser/user-agent of your requests, and the approximate location (such as country or region) derived from your IP address. We process these first-party usage records on our own servers, with no third party involved, both to operate the Service and to understand how it is used (our "first-party product analytics"). Some records (such as logins, uploads and bandwidth) are necessary to run, secure and bill the Service. The rest are used for first-party product analytics on the legitimate-interest basis described in Section 4, and you can opt out of that analytics use at any time in Settings, under Privacy & cookies (we still keep the records necessary to operate and secure the Service).
  • How you found us (signup attribution): when you register or join the waitlist, we record coarse, first-party attribution: the source that referred you (for example a link from a social network, or a campaign tag in the web address) and your approximate country. We use this on the legitimate-interest basis in Section 4 to understand where our users come from and improve our outreach. It holds no precise location and involves no third-party tracker. Our funnel counts (how many people began signing up from a given source) are anonymous: we do not store the email of anyone who starts but does not complete signup.
  • Analytics data (optional, consent-based): if you accept optional analytics, our analytics provider PostHog records how you use the Service (such as pages viewed and features used) to give us richer insight. This is separate from our first-party analytics, is off unless you accept it, and you can withdraw consent at any time. See our Cookie Notice and Section 6 below.
  • Browser-extension data: what our browser extension keeps on your device and what it sends us when you use it are set out in Section 11b.
  • Papers you share: when you share a paper with people, we keep the share: the paper's details, the note and quotes you choose to include, and who it went to (the people you pick and the email addresses you type). We email each recipient separately. If an address has no ReadNTag account, we keep the address and the pending share for at most 30 days, so that the person can sign up and find it, and then delete them. We also keep your sharing settings (who may share with you and what other people see about you), the people you block, and any share you report. If you use Copy link, anyone who has the link can open a page with the paper's details, your note, the quotes you chose and, if your Sharing settings show it, your name. The link works for 30 days unless you turn it off sooner.
  • Library access: if you give us the address of your library's proxy (in Settings, or when our browser extension finds it as you use your library), we keep it in your account so that papers can open through your library.
  • Writing record: when you cite papers with our Google Docs add-in and the writing record is on, we keep each document's title, a random identifier the add-in stores in the document, and which papers from your library it cites, with the dates we saw them, so that a paper can show where you cited it. We never keep the document's text. You can switch the record off, forget a document, or clear the whole record in Settings.
  • Payment data: payments for paid plans are handled by our payment processor (Stripe). We do not store full card numbers; we keep only limited billing records.
  • Waitlist / pre-launch signup: if you ask to be notified about ReadNTag (for example from a coming-soon page) or if signups are temporarily full when you register, we store the email address you provide, together with the coarse source that referred you and your approximate country (see "How you found us" above), so we can invite you when a place opens. You can remove your waitlist entry at any time (see Section 9).

3. How we use your data

We use personal data to: create and secure your account; store and display your library and annotations; provide features you ask for (capture, enrichment, citation, export, sharing); maintain, debug and improve the Service; prevent abuse and fraud; comply with law; notify you when a waitlist place opens if you asked to be added to our waitlist; and, where you have agreed, send you service or marketing communications (which you can opt out of at any time).

We do not train AI models on your content. We do not use your uploaded documents, annotations, notes, or other content you create to train artificial-intelligence or machine-learning models, and we will not do so without first asking for your consent. Section 5a explains where artificial intelligence is used in building, running and supporting the Service, and in the features we offer you.

4. Legal basis / consent

Under the PDPA, we collect, use and disclose your personal data with your consent and for the purposes notified in this Policy. Where the GDPR applies, we rely on: performance of our contract with you (to provide the Service), our legitimate interests (to secure and improve the Service), your consent (e.g. optional third-party analytics or marketing), and legal obligations.

Our legitimate interests include first-party product analytics: measuring how the Service is used, on our own servers and with no third party involved, so that we can improve it. We have weighed this against your interests and rights, keep the data first-party and free of advertising, minimise what we record, and offer an opt-out in Settings, under Privacy & cookies. Optional third-party analytics (PostHog) relies on your consent, not on legitimate interest, and is off until you accept it.

5. Third-party enrichment (what leaves our servers)

To enrich references, the Service may query public scholarly APIs such as Crossref, OpenAlex, Unpaywall and PubMed/arXiv. Only identifiers you already have (such as a DOI or title) are sent; we do not send your account details or private notes. These lookups are server-to-server and degrade silently if a provider is unavailable. This section describes reference enrichment only. Artificial intelligence is covered separately in section 5a.

5a. Artificial intelligence

We use artificial intelligence in two distinct ways, and we want to be plain about both.

To build, run and support the Service. We use AI-assisted development, operations and support tools to write and review code, investigate faults, monitor the Service, and handle support requests. While doing that work, those tools may process data held in the Service, including content you have created. This is the same access our own people have, and it is limited to what is needed to build, operate, secure, debug and support the Service.

To provide features you ask for. Where the Service offers an artificial-intelligence feature, such as summarising a paper, answering questions across your own library, extracting or tagging information, or drafting text from your own notes, the content that feature needs is sent to an artificial-intelligence provider so that the result can be returned to you. We will make clear in the Service which features work this way, and you choose whether to use them.

Which providers. We may use more than one provider of artificial-intelligence models and services, and we may change providers over time. Every commitment in this section applies to each of them. The providers we currently use are named on our Sub-processors page, and we add a new provider there before it starts processing personal data.

We do not train models on your content. We do not use your uploaded documents, annotations, notes or other content you create to train artificial-intelligence or machine-learning models, and we will not change this without asking for your consent first. With each artificial-intelligence provider, we use terms and settings that switch off training on the content we send. Two limits apply under the provider's own terms: it may keep what it receives for a limited period, for example to detect misuse or to meet a legal obligation, and if its systems flag something for safety review, it may analyse that content and use it to improve how it detects misuse of its service, including by training the models it uses for that.

Where our responsibility sits. We choose our providers carefully, we use them only under their published terms and the settings described above, and we remain responsible to you for the providers we engage, as the law requires. What we cannot do is guarantee the internal conduct of another company. If a provider ever failed to meet those terms or settings, we would act promptly to stop the processing, move to another provider where needed, and tell you if the failure affected your data.

Your content remains yours, and you can still remove it. Using an artificial-intelligence feature does not change who owns your content, and it does not affect your right to export or delete it.

6. Sharing & sub-processors

We do not sell your personal data. We use a small number of trusted service providers ("sub-processors") to operate the Service, each only for the purposes described here and under that provider's terms. They fall into these categories:

  • Cloud hosting and storage, for our database and your files, in the United States.
  • Account sign-in and authentication, in the United States.
  • Transactional email, such as verification, password reset and notifications, in the United States.
  • Payment processing for paid plans.
  • Optional product analytics, only if you consent. This is in addition to the first-party analytics we run on our own servers (Section 2); declining it does not limit your use of the Service.
  • Artificial intelligence, for building, operating and supporting the Service, and for artificial-intelligence features where the Service offers them. We do not train models on your content, and training on it is switched off with these providers, within the limits in section 5a.
  • Error monitoring and crash diagnostics. These receive technical error data such as stack traces and the request context when something goes wrong, which can incidentally include limited identifiers; they do not receive your uploaded documents or annotations.

The providers we currently use in each category are named on our Sub-processors page, with what each one does and where it processes data. That page forms part of this Policy, and we update it before a new provider starts processing personal data.

We also disclose personal data to authorities where we are legally required to do so.

Business transfers. If we are involved in a merger, acquisition, financing, reorganisation, sale of assets, or a transfer of all or part of our business, your personal data may be transferred to the successor or acquirer as part of that transaction. We will require the recipient to continue to protect your personal data in a manner consistent with this Policy, and we will notify you of any such transfer and of any material change to how your data is handled.

User-initiated integrations. If you choose to connect an integration (for example, Microsoft OneNote via Microsoft Graph, Notion, or a custom webhook URL you provide), the Service sends the reference or note content you push to that third party at your request. Those third parties process the pushed content under their own terms and privacy policies, and they act as the controller of whatever you send them. You can disconnect an integration at any time in your settings.

Public reading lists, where we offer them, expose only bibliographic fields (title, authors, journal, DOI) of the papers in that list. They never expose your annotations, files, or account information, and only while the share link is active.

6a. Advertising and promotions

The Service may include ads and promotions, on free and paid plans alike. Where we show them, we make them relevant using information we already hold about you as a first party, such as the topics and tags in your library. We do not sell or share your personal data with advertisers, and we do not use third-party advertising or cross-site tracking cookies for this. Advertisers do not receive your personal data; we choose what to show using our own signals. This is separate from the optional product analytics described in Section 2, which we never use for advertising. If and when we offer personalized promotions, you will be able to turn off the personalization in your settings and still see non-personalized ones.

Some content, such as certain conference listings, may be sponsored, meaning a paid placement. We always label sponsored content clearly.

7. Where your data is stored (cross-border)

Your data is hosted in the United States. If we process or back up data outside Singapore, we ensure, as required by the PDPA's Transfer Limitation Obligation, that it receives a standard of protection comparable to the PDPA, through our providers' contractual commitments and our own security controls.

8. Retention

We keep your personal data for as long as your account is active and as needed to provide the Service. When you delete your account, it is deactivated immediately and then permanently erased after a 30-day recovery period. We keep your data during those 30 days only so that an accidental or unauthorized deletion can be reversed using the link we email you; you can also ask us to erase it sooner. After the recovery period we erase your content and personal data from our active systems, except where we must retain limited records to comply with law, resolve disputes, or enforce our agreements. Backups are purged on our routine backup cycle. So that restoring a backup can never bring back a deleted account, we also keep a minimal record of each account deletion outside our backups: a one-way code made from the account's internal ID (never your name, email or content), whether the deletion was asked for, cancelled or completed, and when. We keep it for as long as it is needed to honour the deletion.

We may also delete free accounts that have been inactive for 12 months, so that we do not hold personal data longer than we need. If we do, we email you at least 30 days beforehand, so you can keep the account simply by signing in; if it stays inactive, it is deleted through the same recovery-period process described above.

A share to an address with no ReadNTag account is kept for at most 30 days, and then deleted together with the address. Files you remove from a paper go to Trash, where you can restore them for 30 days; then they are deleted (you can empty Trash sooner).

If you ask us to stop share emails (from one person, or all of them), we keep your address in hashed form for as long as that request stands, also after any account is deleted, so that we can honour it and never email you again. Usage counts from our browser extension (how often its features are used, never what you read) are kept for 13 months and then deleted. So is the record of what you do in the app: each paper's activity history, and how you use the home feed. A sign-up that was started but never finished is deleted after 90 days.

A waitlist email is kept only until you are invited to join or you ask us to remove it, whichever comes first. If you later create an account and then delete it, any waitlist entry for your email is removed as part of that erasure.

9. Your rights

You may, at any time:

  • Access & correct your account data in Settings;
  • Export your data. Download a copy of your references, annotations and account data at any time ("Export my data", a machine-readable file). You can also request a full export that additionally includes your original files (PDFs, attachments and snap images), which we prepare and email you a download link for. The full export is available once per billing cycle and counts toward your monthly download allowance; the metadata export above is always available. If your allowance is used up, or you need a complete copy sooner, contact support@readntag.com and we will provide it free of charge (these usage limits apply only to the self-service export and never restrict your right of access);
  • Delete your account. Request full erasure ("Delete my account");
  • Opt out of first-party analytics in Settings, under Privacy & cookies;
  • Withdraw consent to optional processing (e.g. third-party analytics or marketing);
  • Leave the waitlist. If you are only on our pre-launch waitlist (no account), remove your email at /waitlist/remove, or email us.

To exercise any right not available in-app, contact dpo@readntag.com. We respond within the timeframe required by law.

10. Security

We protect your data with encryption in transit (TLS) and at rest, access controls, hashed passwords, audit logging, and regular dependency and backup hygiene. No method of transmission or storage is ever completely secure, but we work to apply protections appropriate to the sensitivity of the data.

Notifying you of a breach. If we become aware of a security breach that affects your personal data, we will investigate without delay, take steps to contain and remediate it, and inform you and any relevant data-protection authority where the law requires us to. Where the law sets a deadline for that notice, for example the 72-hour window under the GDPR or the timelines under Singapore's Personal Data Protection Act, we will meet it. Our notice will explain, as far as we know it at the time, what happened, the categories of data involved, the likely effect, and the steps you can take to protect yourself.

11. Cookies

We use strictly-necessary cookies and local storage for sign-in and your preferences; these always run. With your consent, we also set an analytics cookie via PostHog to measure how the Service is used. You can give or withdraw that consent at any time from the "Cookie preferences" link in the footer or in Settings, under Privacy & cookies. Our first-party product analytics does not rely on these optional cookies and has its own opt-out in the same place. See our Cookie Notice for full details.

11a. Word and Google Docs add-ins

This section applies when you use one of our add-ins for Microsoft Word or Google Docs, which let you cite from your library while you write. It explains exactly what each one touches.

What the add-in reads. The add-in reads the document you have open, and only that document, so it can find and update the citations and bibliography it has placed there. It does not read your other documents and it does not read your files in any connected storage service.

What leaves your document. When a citation is inserted or a bibliography is rebuilt, the citation details for the works being cited are sent to us to be formatted. That is the reference information itself, such as authors, title, journal, year, and page numbers. The rest of your document is not sent to us, and we do not store the text you write.

The writing record (Google Docs). When the writing record is on (it is on unless you switch it off in Settings › Writing record), the Google Docs add-in also sends us the document's title and a random identifier it keeps in the document, so that your library can show where you cited each paper. It never sends the text you write. You can switch it off or clear it at any time.

Where your connection key is kept. Connecting an add-in to your account creates a key. In Word it is held in the add-in's own storage in that browser or application profile. In Google Docs it is held against your Google account, not inside the document. In both cases the key is never written into the document, so sharing the document with a co-author does not share your key or give anyone access to your library.

Working with co-authors. Because each citation carries its own reference details inside the document, a co-author can reformat citations and rebuild the bibliography without a ReadNTag account. When they do, only those reference details are sent to us for formatting. They are not associated with an account, and we do not create one for them.

Google user data (Limited Use). ReadNTag's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements. We use Google user data only to provide the features described here. We do not transfer it to others except as needed to provide those features, to comply with applicable law, or as part of a merger or acquisition. We do not use it for advertising, and we do not allow humans to read it except with your explicit consent for support you have asked for, where required by law, or on data that has been aggregated and made anonymous.

11b. Browser extension

This section applies when you use our browser extension for Chrome, Firefox and other browsers, which saves papers and web pages to your library. It explains what the extension reads, what it keeps on your device and what it sends us.

What the extension reads. It reads a page when you open ReadNTag on it, save it or snap part of it. While its side panel is open, it also reads, as you move between tabs, research sites, PDFs and the sites you added to Always allow, and it waits on other pages until you ask. When the PDF reader is on, it also reads a research PDF you open, inside your browser, to show it in the ReadNTag reader. It never reads your Blocked sites. On the webmail, payment and banking sites on our list (such as Gmail, Outlook and PayPal) it reads nothing until you confirm, and from any page with a password or payment field it sends nothing until you confirm.

What it sends us. When you open ReadNTag on a page, it sends that page's address and DOI to your ReadNTag account to check your library and suggest topics, and asks us for any notice about the extension, with its version number. Nothing is saved until you click Save. When you save, it sends the page's bibliographic details (such as title, authors, journal and DOI), the topics, projects and note you chose, the quotes you captured with the page's address and where they sit on the page, and the files you ticked, which it downloads from the publisher using your own browser session. A snap sends only the area you selected; saving a figure or a table sends only that figure or table.

What it does by itself. Three features work while the panel is closed, and each has a switch in the extension's Settings. Badges on research search pages: the extension downloads a list of papers that carry a retraction or a similar notice (public data from Crossref) and a coded copy of the papers in your library, and compares them with the search results on your device; the search page is not sent to us. Your highlights on pages you return to: the extension keeps a coded list of the pages where you saved quotes, and when you open one of them again it asks us for that page's quotes by its code, not by its address. Research PDFs in the ReadNTag reader: when this setting is on, a research PDF you open in your browser opens in the ReadNTag reader instead of the browser's PDF viewer. The file stays in your browser: nothing is uploaded and the PDF's address is not sent to us until you click Save to library, and every preview offers "Open in the browser's viewer". To show whether the paper is already in your library, and its title and authors from Crossref, the reader sends us the DOI printed in the PDF, never the file or its address. Depending on when you installed the extension, this setting starts on, starts off, or asks you the first time you open a research PDF. The first-run screen and the extension's Settings always show its state, and your own choice always wins.

What it keeps on your device. Your connection key, your settings, the coded lists above, and counts of feature use waiting to be sent. Saves and changes that could not reach us (the page's address and details, your note, quotes, topics and projects), until they are sent: the extension tries again by itself, deletes what it still could not send after a week at most, and tells you. While a PDF is open in the reader preview, the extension keeps the file on your device: it deletes it when you save the paper, return to the browser's viewer or close the tab, and deletes any file it still holds after two hours or when your browser next starts. Your connection key goes only to ReadNTag, with each request, and is never written into a page.

Error reports and feature counts. If something goes wrong, the extension sends us a report with technical details only (never page content, addresses or anything from your library), which we pass to our error-monitoring provider; you can switch this off in the extension's Settings. It counts how often its features are used (such as saves and snaps) on our own servers, unless you turned first-party analytics off in your account settings. It never loads a third-party analytics script. In Firefox, both stay off until you allow them when Firefox asks.

Sharing and library access. When you share a paper from the extension, the addresses and the note you type are sent to us as described in Section 2 (Papers you share); in Firefox, Firefox asks you before the extension sends them. If you use a library proxy, the extension can open publisher pages through it, and the proxy address is kept in your account (Section 2, Library access).

Its permissions. The extension asks for access to all websites so that it can read the page you open it on, keep its side panel working as you move between tabs, show badges on research search pages, take snaps and, when you turn it on, open research PDFs in the ReadNTag reader. It uses that access only as described in this section.

12. Children

The Service is not directed to children under 13 (or the minimum age in your jurisdiction). Where the EU or UK GDPR applies, you must be at least 16, or the minimum digital-consent age set by your member state, to use the Service without the consent of a parent or guardian. We do not knowingly collect personal data from children below the applicable age. If you believe we have, contact dpo@readntag.com and we will delete it.

13. Changes

We may update this Policy. Material changes will be notified in-app or by email. The "Effective date" above reflects the latest version.

14. Contact & complaints

Questions or requests: dpo@readntag.com. If you are in Singapore and are not satisfied with our response, you may contact the Personal Data Protection Commission (PDPC).

ReadNTag · Scinova Group LLP · Singapore. Questions: dpo@readntag.com.