Privacy Policy
Effective: March 22, 2026
Last updated: November 18, 2026
This Privacy Policy describes how Navikaa ("Navikaa", "we", "us", or "our") collects, uses, stores, and shares information about you when you use our AI-powered product management platform (the "Service"). Please read this policy carefully. By using the Service, you agree to the practices described here.
1. Who We Are
Navikaa is an AI-powered product platform designed to help product builders and teams create PRDs, user stories, competitive analyses, OKRs, release notes, and other product artefacts.
Navikaa is operated by Amit Vijayan (Kerala, India), pending incorporation. For the purposes of applicable data protection law, he is the data controller of your personal data, and he is the person you contact about it. When a company is incorporated to operate Navikaa, this policy will be updated to name it as the controller and you will be notified under Section 18.
Contact: privacy@navikaa.ai
2. Data We Collect
2.1 Data you provide directly
- Account data: Name, email address, and password (hashed by our authentication provider — never stored in plain text) when you register.
- Profile data: Job title, company name, and preferences you set in your account.
- Conversation content: Prompts, instructions, and text you enter into AI agents.
- Uploaded documents: Files you attach to conversations for analysis or context. These are used only to provide the Service to you.
- Feedback and support: Messages you send us via email or in-product feedback tools.
2.2 Data collected automatically
- Usage data: Which agents and features you use, frequency, and session duration.
- Technical data: IP address, browser type, operating system, and device identifiers.
- Log data: Server access logs, error events, and performance data.
- Session recordings: See Section 12 for how analytics and session recording actually work, including when they load and what declining does. Section 7 covers error monitoring (Sentry), which records no sessions.
2.3 Data from third-party integrations
When you connect external tools (e.g., Atlassian Jira and Confluence, or Notion), we access only the data necessary to fulfil the specific agent task you have requested. See Section 6 for details on each integration.
3. How We Use Your Data
- To provide the Service: Process your requests, run AI agents, and deliver outputs to you.
- To maintain and improve the Service: Monitor performance, diagnose errors, and develop new features.
- To manage your account: Authentication and account administration.
- To communicate with you: Send service updates, security alerts, and journey share notifications. We do not send marketing emails without your explicit consent.
- To ensure safety and compliance: Detect abuse, enforce our Terms of Service, and meet legal obligations.
- To produce anonymised analytics: Aggregate, non-identifiable usage statistics to understand how the Service is used overall.
4. Legal Basis for Processing (GDPR)
For users in the EU/EEA, we process your personal data on the following legal bases under GDPR:
- Contractual necessity (Art. 6(1)(b)): Account management, AI agent execution, and delivering outputs.
- Legitimate interests (Art. 6(1)(f)): Security monitoring, error logging, session recording to diagnose crashes (where our interests do not override your rights).
- Legal obligation (Art. 6(1)(c)): Where we are required by law to retain or disclose data.
- Consent (Art. 6(1)(a)): For optional communications such as product newsletters. You may withdraw consent at any time.
5. AI Agents and Your Content
- We do not train AI models on your content. Your prompts, documents, and outputs are never used to fine-tune, train, or improve our own models — and, per our AI providers' published API data policies, are not used to train theirs.
- Content is processed in-context only. Your input is sent to our AI providers solely to process your request: Anthropic for generation and for the web searches used to fact-check your plan, OpenAI for document embeddings, and Together AI only as a standby if Anthropic is unavailable (see Section 8). These providers process data only to complete your request and do not use API-submitted data to train their models by default.
- Outputs are yours. You retain full ownership of outputs generated for you by Navikaa's agents.
- Honesty Protocol. When an agent cannot ground a response in your live data, it explicitly states this. We never silently substitute training data for live data.
- You can delete your content. Conversations and documents can be deleted from your account at any time via Settings. If you ask us to delete your whole account, the request is held for a 30-day grace period — so that you can cancel it if you change your mind — and your data is then erased. You can cancel at any point during those 30 days.
6. Third-Party Integrations
6.1 Atlassian (Jira + Confluence)
- What we access: Jira issues, sprints, backlog data, and Confluence pages — only when you invoke an agent that requires this data, or when you explicitly trigger a push action.
- What we write: Only on your explicit action. Navikaa never autonomously writes to your Atlassian tools.
- What we store: OAuth tokens (encrypted with AES-256-GCM at rest). Minimal page metadata (title, space name) for display. We do not store the content of your Jira issues or Confluence pages.
- Revocation: Disconnect at any time from Settings → Connections. This immediately invalidates our stored tokens.
6.2 Notion
- What we access: Notion page content and workspace metadata — only when you invoke an agent that reads from Notion, or when you explicitly trigger a push action to write agent output to a Notion page.
- What we write: Only on your explicit action (the Push to Notion button). Navikaa appends content to the Notion page you designate. We never autonomously write to your Notion workspace.
- What we store: OAuth access token (encrypted with AES-256-GCM at rest). Saved destination metadata (page ID, page title, workspace name) for the Notion Push Destinations feature. We do not store the content of your Notion pages.
- Revocation: Disconnect at any time from Settings → Connections. This immediately invalidates our stored token. You may also revoke access from Notion's own settings at notion.so/settings/my-connections.
6.3 Other integrations
The same principles apply to all integrations: minimum necessary access, explicit user action for writes, no content storage beyond what is required to render the UI, and immediate revocability.
7. Error Monitoring
We use Sentry (Sentry.io, Inc., USA) for error monitoring and application reliability. You should be aware of the following:
- Error reports: When an error occurs, Sentry receives a diagnostic report — the error message, a stack trace, and the technical context needed to locate the fault. We have disabled the sending of personal data by default, so these reports do not include your IP address, your request headers, or your session cookies.
- What does identify the report: so that we can tell which account hit a fault, an error report carries your internal account identifier — a random id we generate, not your name or your email address. Sentry also derives an approximate location (city and country) when it receives the report, from the network connection it arrives on. We do not send your IP address, and the stored report records none — the location is derived on arrival.
- Performance traces: Approximately 10% of requests are sampled as performance traces — timing data for page loads and server operations — which we use to find slow paths in the Service.
- No session recording. Sentry's session replay feature is switched off, for ordinary sessions and for sessions where an error occurs alike. Sentry captures no recording of your use of the application: no mouse movements, no clicks, no page interactions.
- Data transfer: Sentry servers are located in the United States. Transfers from the EU/EEA are covered by Standard Contractual Clauses.
- Retention: Sentry retains error and performance data for 90 days under its default plan setting.
8. Subprocessors
We use the following third-party subprocessors to operate the Service. All are contractually bound to appropriate data protection standards.
| Subprocessor | Purpose | Location |
|---|---|---|
| Supabase | Authentication, database, and storage | USA / EU |
| Anthropic | Primary AI model inference (Claude). Anthropic states that conversation content — prompts and outputs — is not retained by default, and that retained data is never used for model training without express permission. A 30-day retention requirement applies to certain of Anthropic's models; Navikaa does not use any of them. See Anthropic's API data-retention documentation. | USA |
| Vercel | Application hosting and global edge network | USA / Global CDN |
| Together AI | Standby AI model inference, used only if our primary provider is unavailable. Our Together account is configured for zero data retention: Together does not store the prompts we send or the responses its models return, and the forwarding of prompts to third-party “passthrough” providers is off. Using data to train models is off by default and we have not enabled it. See Together's privacy and security documentation. | USA |
| OpenAI | Document embeddings for search (optional feature). OpenAI states that data sent to its API is not used to train or improve its models unless you explicitly opt in, and that abuse-monitoring logs are retained for up to 30 days unless longer retention is required by law. See OpenAI's API data-usage documentation. | USA |
| Resend | Transactional email (e.g., journey share notifications) | USA |
| Sentry | Error monitoring and performance traces — no session replay (see Section 7) | USA |
| Atlassian | OAuth provider for Jira + Confluence integrations (when connected) | USA / Australia |
| Notion | OAuth provider for Notion integration — page reads and push writes (when connected) | USA |
| Google Analytics 4 (Google LLC) | Website usage analytics — pages viewed, traffic sources, aggregate engagement. Loads only after you accept analytics cookies, and on our public pages only — it is not loaded inside the signed-in app, on shared links, or on any sign-in, password or verification page. See the note in Section 12 for the one limit on that. | USA |
| Microsoft Clarity (Microsoft Corporation) | Website analytics — anonymised session replays and heatmaps, with sensitive inputs masked. Loads only after you accept analytics cookies, and on our public pages only — it is not loaded inside the signed-in app, on shared links, or on any sign-in, password or verification page. See the note in Section 12 for the one limit on that. | USA (Azure) |
| PostHog (PostHog Inc.) | PostHog for cookieless, aggregated product analytics — no cookies, no session recording, user id hashed before events are sent. Always active (not routed through the analytics-consent banner) because it sets no cookies and no directly-identifying data. | USA |
| Langfuse (Langfuse GmbH) | AI observability — traces, cost and latency for AI generations. Receives AI output previews (first 500 characters of a generation), a pseudonymised user identifier, the model name, token counts and computed cost. Never your email address or name. | EU (cloud.langfuse.com) |
On the pseudonymised identifier: we never send your email address or your name to either analytics service. What we send instead is a one-way pseudonym derived from your internal account id, and it is different for each service, so the same person cannot be correlated between them. For Langfuse it is a keyed pseudonym (HMAC-SHA256 under a key held only by us); for PostHog it is a SHA-256 hash. Both arepseudonymised, not anonymous: because we hold the key and the mapping that could re-attribute them, they remain personal data under GDPR (Recital 26) and the rights in Section 9 apply to them. What the pseudonym guarantees is that the third party cannot reverse it on its own.
Langfuse platform change: From 16 November 2026, Langfuse Cloud's terms and data processing agreement are provided by ClickHouse, which acquired Langfuse. This does not change what we send or why.
We will update this table when we add or change subprocessors and notify users of material changes.
9. Data Retention
| Data Type | Retention Period | Basis |
|---|---|---|
| Account data | Duration of account + 30 days after deletion | Contractual |
| Conversation history | Until you delete or account is closed | User control |
| Uploaded documents | Until you delete or account is closed | User control |
| Integration OAuth tokens | Until you disconnect or delete your account | Contractual |
| Server access logs | 90 days | Legitimate interests (security) |
| Sentry error events and performance traces | 90 days (Sentry default) | Legitimate interests (reliability) |
| Anonymised usage analytics | Up to 3 years | Legitimate interests (product improvement) |
| Support correspondence | 3 years from last interaction | Legitimate interests |
When you ask us to delete your account, the request is held for a 30-day grace period during which you can cancel it. At the end of that period your personal data is deleted or anonymised, except where a longer retention period is required by law.
10. Data Sharing and Disclosure
We do not sell, rent, or trade your personal data. We may share data only in these limited circumstances:
- Subprocessors: As listed in Section 8, solely to provide the Service.
- Legal requirements: Where required by law, court order, or to protect rights, property, or safety.
- Business transfers: In the event of a merger, acquisition, or sale of assets, your data may be transferred. We will notify you in advance.
- With your consent: For any other purpose with your explicit consent.
10.1 Internal access to your content
Navikaa is a small operation, and we would rather tell you plainly how internal access works than leave you to assume.
- Administrative access is limited to the founder. No one else holds an administrative account on the Service.
- We do not open your content without your permission. We will not read your conversations, documents, or generated artefacts for curiosity, for product research, or to build features. The two exceptions are investigating suspected abuse of the Service, and meeting a legal obligation — and we keep those to the minimum the situation requires.
- Other users cannot see your content. Your work is visible only to you unless you choose to share it — by sharing a journey or creating a public link. Where you do share, anyone holding that link can view what you shared, so treat a public link as public.
11. Security
We implement the following measures to protect your data:
- All data in transit is encrypted using TLS (HTTPS enforced for all connections).
- MCP integration credentials (OAuth tokens, API keys) are encrypted at rest using AES-256-GCM before storage.
- Passwords are hashed by our authentication provider (Supabase). We never store plain-text passwords.
- Security headers are applied to all responses: X-Frame-Options, X-Content-Type-Options, Referrer-Policy, Strict-Transport-Security (HSTS), and Permissions-Policy.
- MFA (two-factor authentication via TOTP) is available for all accounts — you can enable it in Settings, and once enabled you are asked for your authenticator code when you sign in. Mandatory MFA on sensitive operations is being rolled out and is not yet in force across the whole application.
- We conduct regular internal security reviews to identify and address vulnerabilities.
Despite these measures, no system is completely secure. In the event of a data breach affecting your rights and freedoms, we will notify you and the relevant regulatory authority as required by applicable law.
To report a security vulnerability or concern, please contact us atsecurity@navikaa.ai.We take all security reports seriously and will respond promptly.
12. Cookies and Tracking
Navikaa uses the following types of cookies:
- Strictly necessary cookies: Session authentication tokens required for you to log in and use the Service. These cannot be disabled without breaking the Service.
- Preference cookies: Store your UI preferences (e.g., sidebar state). Functional only — do not track you across other sites.
- Analytics cookies (optional — consent required): Set only after you accept analytics in the consent banner. If you decline, no analytics script loads and none of these cookies are set. If you accepted earlier in the same visit and then declined, we reload the page for you so that the scripts already running stop — see the note at the end of this section.
- Google Analytics 4:
_ga,_gid(and_ga_<id>) — distinguish visitors and measure aggregate usage such as pages viewed, sessions, and traffic sources. They do not identify you by name. - Microsoft Clarity:
_clck,_clsk,CLID— associate a session for anonymised behavioural analytics, heatmaps, and session replays. Sensitive form inputs are masked in recordings.
- Google Analytics 4:
Analytics load ONLY after you grant consent, and never before. You can decline at any time in the consent banner. We then delete the analytics cookies above where your browser lets us — a cookie set on a domain we cannot target, or flagged HttpOnly, has to expire on its own — and we load no analytics script again. Withdrawing consent takes effect immediately: we reload the page for you, which stops the scripts that were already running — you do not have to do anything. We are deliberately precise rather than absolute: a request already in flight at that moment may still complete, and a cookie we cannot target or that is flagged HttpOnly has to expire on its own. We do not use advertising cookies, participate in ad networks, or engage in behavioural advertising of any kind.
Note on session recordings: Only one of our tools records sessions at all: Microsoft Clarity (analytics, consent-gated), which captures anonymised interaction replays with sensitive inputs such as passwords masked. We load it on our public pages only — it is not loaded inside the signed-in app, on shared links, or on any sign-in, password or verification page. The same scope applies to Google Analytics.
How we keep that true, and the one limit we will not paper over. Those are rules about loading, applied when a page loads — so the question is whether you can reach one of the pages above without a page load. Two things make that hard: a page whose first address was one of those pages never starts an analytics script at all, even if you move to a public page and come back; and the links from our public pages into them are deliberately full page loads, which re-apply the rules. An automated check runs on every change, walking outwards from every public page we serve, and fails the build if one of those links turns back into an in-page navigation. The limit: that check reads our code, so it covers links we write down — it cannot see one assembled at run time from a variable. We would rather say that than claim a guarantee we cannot inspect. If you want certainty after withdrawing consent, the reload we perform for you is what ends any script that was running. Sentry (Section 7, diagnostic) records no sessions: its session replay feature is switched off. Separately, we use PostHog for cookieless, aggregated product analytics — it sets no cookies, records no sessions, and hashes your user id before any event is sent; it is not gated by the analytics-consent banner because it stores no cookies and no directly-identifying data.
13. International Data Transfers
Most of our subprocessors are located in the United States. When we transfer personal data from the EU/EEA to the US, we rely on Standard Contractual Clauses (SCCs) approved by the European Commission, or the EU–US Data Privacy Framework where applicable.
Specifically, when you accept analytics, Microsoft Clarity and Google Analytics process data on servers in the United States. These transfers rely on the Standard Contractual Clauses incorporated in Google's and Microsoft's data-processing terms, extended by the UK International Data Transfer Addendum for UK visitors and the Swiss amendments for Swiss visitors.
For transfers of Indian personal data outside India, we comply with the conditions set forth under the Digital Personal Data Protection Act (DPDPA) 2023.
14. Your Rights (GDPR — EU/EEA Users)
If you are located in the EU or EEA, you have the following rights under GDPR:
- Right of access (Art. 15): Request a copy of the personal data we hold about you.
- Right to rectification (Art. 16): Request correction of inaccurate or incomplete data.
- Right to erasure (Art. 17): Request deletion of your data, subject to legal retention obligations.
- Right to restriction (Art. 18): Request that we limit processing of your data.
- Right to data portability (Art. 20): Receive your data in a structured, machine-readable format.
- Right to object (Art. 21): Object to processing based on legitimate interests.
- Right to withdraw consent: Where processing is based on consent, withdraw it at any time without affecting prior processing.
To exercise any of these rights, email privacy@navikaa.ai. We will respond within 30 days. You also have the right to lodge a complaint with your local supervisory authority.
15. Your Rights (CCPA/CPRA — California Users)
- Right to know: Know what personal information we collect, use, and disclose. We do not sell your data.
- Right to delete: Request deletion of personal information, subject to certain exceptions.
- Right to correct: Request correction of inaccurate personal information.
- Right to opt out of sale/sharing: We do not sell or share personal information for cross-context behavioural advertising.
- Right to non-discrimination: We will not discriminate against you for exercising your CCPA rights.
To submit a CCPA request, email privacy@navikaa.ai with subject "CCPA Request". We will respond within 45 days.
16. Your Rights (DPDPA — India)
Under India's Digital Personal Data Protection Act (DPDPA) 2023, you have the right to:
- Information: Obtain information about the personal data we process and the purposes.
- Correction and erasure: Request correction of inaccurate data and erasure of data no longer needed.
- Grievance redressal: Have grievances addressed within 30 days.
- Nomination: Nominate another individual to exercise your rights in the event of death or incapacity.
16.1 Notice to Indian users
This notice is given to meet Rule 3 of the Digital Personal Data Protection Rules, 2025, which requires a notice that stands on its own, is written in plain language, and tells you what data is collected, why, how to withdraw consent, how to exercise your rights, and how to complain. The Rules were notified on 13 November 2025; Rule 3 commences eighteen months later (around 13 May 2027). We are following it now rather than waiting.
- What we collect, itemised: your name and email address; your password (stored only as a hash by our authentication provider); the product content you create or upload (prompts, documents, journeys, generated artefacts); your UI preferences; and technical logs including IP address and browser type. Section 2 lists these in full.
- Why we collect it, itemised: to create and secure your account; to run the AI agents that produce your artefacts; to store your work so you can return to it; to keep the Service reliable and to diagnose faults; and, only if you accept analytics cookies, to understand aggregate usage of our public pages. Section 3 sets out each purpose.
- How to withdraw your consent, as easily as you gave it: where we rely on your consent, you can withdraw it by the same means you gave it. Analytics consent can be changed at any time from the cookie banner link in the site footer, with no account needed and no explanation required. Connected integrations can be disconnected in Settings. Withdrawing consent does not affect processing we are required to carry out for the contract or by law.
- How to exercise your rights: email privacy@navikaa.ai. You can also delete your content and your account yourself from Settings at any time — see Section 5.
- How to complain: write to us first at privacy@navikaa.ai; we respond to grievances within 30 days. If you are not satisfied, you may complain to the Data Protection Board of India, which is the authority established under the DPDP Act to hear complaints from Data Principals.
Contact: privacy@navikaa.ai
17. Children's Privacy
The Service is not directed to individuals under the age of 18. We do not knowingly collect personal data from minors. If you believe we have inadvertently collected such data, please contact us immediately and we will delete it.
18. Changes to This Policy
When we make material changes to this policy, we will:
- Update the "Last updated" date at the top of this page.
- Display an in-app notification for active users.
- For material changes affecting how we process your data, provide at least 14 days' notice before the changes take effect.
19. Contact Us
For privacy questions, data subject requests, or any concerns about how we handle your data:
- Email: privacy@navikaa.ai
- Response time: We aim to acknowledge requests within 2 business days and resolve them within 30 days.