Privacy
About this policy
This policy explains the data Nerva processes when you visit the website, contact the team, or use documents, the web workspace and connected services. Send data questions through the contact form. Do not include passwords, access tokens or complete working documents unless they are necessary to address the request.
The organisation connecting Nerva determines its users, permitted connections and documents to process. It should ensure an appropriate basis for providing information about employees, suppliers and other people. Responsibilities for organisational data are clarified in the terms of that organisation's setup.
Information Nerva receives
- Website enquiries: company or venue name, the contact you provide, your needs, language, time, handling status and team notes.
- Account details: login, display name, email, role, organisation, permitted connections and session information.
- Working documents: uploaded files, filenames and technical details, recognised text, document details, line items, quantities, prices, taxes, catalogue matches, corrections and processing history.
- Connections: catalogues, suppliers, warehouses, accounts, figures and records available to the user within the connected system's settings.
- Telegram and the AI assistant: user and chat identifiers, account links, submitted materials, selected preferences, questions, answers and the context of reports used.
Avoid uploading unnecessary personal data. Where possible, remove details that are not needed for processing before uploading a document, while preserving information required for its accounting purpose.
Why the data is processed
Data supports responses to enquiries, access, document recognition and review, product matching, integrations, reports, support and protection against abuse. Operation history helps explain changes and investigate errors.
The legal basis depends on the relationship and applicable law: delivery of agreed services, steps you request, legitimate interests in supporting and protecting the service, legal obligations, or consent where required. Without information needed for login, an enquiry or a document, the corresponding function may be unavailable.
Recognition, AI and learning from matches
Recognition can send file contents, page images and text to the configured AI provider. Product selection can send document lines and catalogue candidates; embedding generation can send item text. The analytical assistant uses your question and information retrieved through permitted tools. Supported providers include OpenAI.
Relevant Responses API requests use store: false. This is not a guarantee
that a provider performs no technical retention: its terms and the
connection settings govern that processing.
Nerva can also use user-confirmed matches to train internal candidate ranking within an organisation and connection. This incorporates its catalogue and previous decisions. An automated suggestion is not itself a human-confirmed training example. Ask your organisation's administrator about the applicable settings.
Other recipients and services
Access depends on user roles and organisational permissions. Functions can involve infrastructure, AI and service-message providers, Telegram and the connected Syrve system. When enquiry notifications are enabled, the company name, contact and enquiry text are sent to the team's configured Telegram chat.
The information transferred depends on enabled functions. External services may process data in other countries; actual processing locations and applicable arrangements should be clarified for the specific connection. Following an external link brings you under that service's own policy.
Cookies and website statistics
Login and protected actions use technical cookies for sessions, protection against forged requests, email login and address changes. Blocking them may prevent the workspace from functioning.
First-party website statistics count events by day, page, language and section without a visitor profile in the analytics table. Statistics requests omit cookies, and the client respects Do Not Track. To limit spam, the server processes the client address and derives a hashed rate-limit key. Infrastructure may also process technical connection logs.
Retention and deletion
Original documents are held in private file storage; structured data is held in the service database with access separation. Retention needs depend on the purpose, relationship duration, accounting obligations and the need to address an enquiry or dispute. Expiry of web access is not itself an instruction to delete documents.
Administrative deletion of an organisation or connection covers related data within the confirmed operation; file cleanup may complete separately. Deleting a user does not delete the organisation's business documents. A record of the deletion action may remain. Cleanup periods in external services and backup copies require separate clarification.
Your requests
Depending on applicable law, you may request access, correction, deletion, restriction or portability, object to processing, withdraw consent where it is the basis, and contact the competent authority. Withdrawal does not change the lawfulness of earlier consent-based processing.
Use the contact form, naming your organisation and describing the request. Protecting data may require verification of your identity and authority. Requests involving working documents are also coordinated with an authorised organisational representative. This policy is updated as the service changes; check the arrangements for your connection before transferring data.