This Privacy Policy explains what personal data Segmova Studio ("Segmova," "we") collects when you use the Service, how we use it, and your choices. It applies to your use of Segmova Studio, currently offered as a private beta.
Account information (your email address and authentication credentials); the content you upload or create — documents, translations, Translation Memory entries, Glossary terms, and the reference files and links you attach to a project (see "Documents and Translation Data" below for how each is processed); if you use Team workspaces, the organization, membership, task and notification data described in "Team Workspaces" below; your acceptance of these Terms and this Policy, with a timestamp; data cached locally in your browser; and basic technical logs needed to operate and secure the Service. We don't collect payment information, as the Service is currently free.
Separately from any account: if you submit the Early Access or Contact form on our website, we collect the name, email address, role and message you provide, along with the language you were reading, which form and call-to-action you used, and a record of the consent you gave together with the version of this Policy in effect at that moment. You don't need an account to use those forms and we don't create one for you. We don't store your IP address, and we don't attach any advertising or cross-site tracking identifier to these submissions.
To create and secure your account and authenticate you; to provide the Service's features, including running Team workspaces — showing members to one another, assigning and tracking work, and notifying people about it; to keep a record of your acceptance of these Terms and this Policy; to communicate with you about your account; to answer the Early Access or Contact requests you send us and to run the Early Access programme; and to keep the Service secure and reliable.
How your document is processed depends on its format. PDF, DOCX, XLSX, JSON, and XML documents are uploaded to our cloud storage provider and parsed and segmented on our own servers; the parsed content and segment text are stored in our database, scoped to your account. HTML, SRT, XLIFF, and TXT documents are parsed and segmented in your own browser; the original file and the resulting segments are then uploaded and saved to your account in the same way, so your work isn't lost if you switch browsers or devices.
For documents stored in your account: access is enforced so that only your account can read or manage them, except that a document in a Team project is also available to that project's members as described in "Team Workspaces" below; they remain stored for as long as your account and that content stay active, and removing a document deletes it, including its stored file. An import that fails or never completes is automatically cleaned up — its stored file and record are deleted — after a bounded period if you don't retry it; a completed document is not deleted on any timer.
A project can also hold reference material that isn't a document — for example a style guide, a screenshot or a client brief — attached as an uploaded file or as a link. An uploaded reference file is stored with our cloud storage provider in a private location scoped to your account, and only your account can download it. A link is stored as text only: our servers never open or fetch it. Reference material is never parsed, segmented, added to Translation Memory, counted against your word quota, or sent to a machine translation or AI provider. Removing a reference deletes it, including its stored file, and deleting a project deletes all of its reference files.
Your Translation Memory entries and Glossary terms are stored in our database, scoped to your account, and private to you — not shared or pooled with other users. Automatically saving a confirmed segment to Translation Memory is on by default; you can turn this off at any time from Translation Memory settings within Studio, so future confirmations are no longer saved there.
Your browser keeps a temporary working copy of the document you currently have open, along with some interface and feature preferences, so your work survives a page refresh. This is not where your documents are kept: every document format, together with your Translation Memory and Glossary data, is stored server-side in your account, as described above. If a document can't be saved to your account, Studio tells you so, and until it is, that working copy is the only one — export your translation if you need to keep it.
API keys for AI providers are never written to your browser's storage. A key you enter is held in memory for the current session only, and you'll need to enter it again after a refresh. Clearing your browser storage removes the working copy and your preferences, not the documents saved to your account.
OCR runs on our own infrastructure. We do not use your content to train any model we operate. If you enable machine translation or AI-assisted segmentation with your own Anthropic or OpenAI API key, the relevant text and your key are sent to our servers, which forward the request to that provider and return its result. Your key is used only for that request: we don't store it in our database, write it to our logs, or keep it once the request completes. AI-assisted segmentation is available only for HTML/HTM imports; PDF, DOCX, XLSX, JSON, and XML files are never sent to an AI provider for segmentation. Handling of data sent to an AI provider on your behalf is governed by that provider's own terms and privacy policy.
We use Supabase for authentication, database, and file storage, and Vercel for hosting. We share data with these providers only as needed to operate the Service, and with an AI provider only when you personally enable and configure that feature. We don't sell your data.
We also use Sentry, hosted in the European Union, to be told when the Service fails. What it receives is deliberately narrow: the error message, the technical stack trace, and the page the failure happened on. It does not receive your IP address, your cookies, or the contents of your requests, and screen recording is not enabled — your documents and the translation grid are never captured. Before any report leaves our servers or your browser, we strip anything matching an API key, an access token, or an email address. Our own source code is not shared with Sentry.
One honest limitation, since it is easier to state than to guarantee away: an error message written by the software can occasionally quote a fragment of the text being processed. We minimise this by collecting as little as possible in the first place, but we can't promise a stray fragment never appears in a technical report. If that matters for a particular document, translate it with error reporting concerns in mind and contact us at support@segmova.com.
We retain your account, documents, project reference files, Translation Memory, and Glossary data for as long as your account remains active. You can permanently delete your account at any time from within the product; if you're unable to, contact support@segmova.com. Deleting your account removes your credentials, profile, documents and project reference files (including their stored files), Translation Memory, Glossary entries, and consent records together, and can't be undone. Separately, an import that fails or never completes is automatically removed after a bounded period if you don't retry it, as described in "Documents and Translation Data" above.
Early Access and Contact form submissions are kept only for as long as we need them to answer you or to run the Early Access programme, and are deleted once that purpose ends. Because these submissions aren't tied to an account, deleting an account doesn't remove them — to have yours deleted sooner, email support@segmova.com from the address you submitted.
We use account-scoped access controls and delegate authentication to Supabase Auth, so we never handle your password directly. No system is completely secure, and we can't guarantee against every risk.
Depending on where you live, you may have rights — for example under the EU's GDPR or Turkey's KVKK — to access, correct, export, or delete your personal data, and to object to certain processing. To exercise any of these rights, contact support@segmova.com.
A manager can set up a Team organization, invite people to it, run Team projects and assign work. If you use Teams — as a manager or as an invited translator or reviewer — we store in our database: the organizations you belong to and your role in each; the Team projects you are a member of and your role there (manager, translator or reviewer); tasks — which document, which step (translation or review), who it is assigned to, its deadline, and its status as it moves from assigned through submission and review to completion; invitations — the invited email address, the role offered, who sent it, its status and its expiry; and whether your account has Team Management access. We keep only a hash of an invitation's secret link, not the link itself, and we don't email invitations — the person who creates one shares the link. When you sign in, a functional cookie records whether you chose Team or translator mode.
Inside a Team, some of your data is visible to other members by design. Members of the same organization or project can see each other's email address and role. A document in a Team project can be opened by that project's managers and by the members assigned a task on it, and the member assigned a step can edit it while that step is open. Managers see who is assigned what, task statuses and deadlines, and each document's progress. Access is checked on every request, so removing a member or reassigning a task takes effect straight away.
Translation Memory and Glossary data stay private to each account in a Team too; they are not shared with other members. A segment you confirm while working in a Team document is saved to your own Translation Memory unless you have turned automatic saving off. Team notifications — for example that a task was assigned to you, that a translation is ready for review, or that work is done — are shown in the app only, not sent by email. Each records the project, the task and who triggered it; notifications older than 90 days are deleted automatically the next time you view your notifications.
Deleting your account also deletes your organization and project memberships, the tasks assigned to you, the invitations you sent, and the notifications addressed to you. If you created a Team organization, it is deleted too — even if other members were also made owners — together with all of its memberships and invitations. Team projects in it that other managers created are not deleted, but every member loses access to them and they remain only with the manager who created them. Team projects you created and documents you uploaded are deleted like any other project or document of yours, together with their tasks and notifications — and with them the translation and review work other members did in those documents. Documents that other members uploaded to a project you created are not deleted: they are detached from the project and its tasks, and stay in the account of the member who uploaded them.
The Service is intended for professional and business use and is not directed at children. We don't knowingly collect personal data from children.
Last updated: 30 September 2026. We may update this Privacy Policy from time to time. A material update carries a new version number, and you'll be asked to accept it before continuing to use the Service. Privacy questions or requests can be sent to support@segmova.com.