Skip to content

Legal · last updated 18 October 2026

Privacy Policy

Zenith is operated by Marceli Pawliński. This policy explains what data Zenith collects, why, and the choices you have.

Who is responsible

Marceli Pawliński is the data controller for Zenith. For any privacy question or request, contact support@marcelipawlinski.co.uk.

Data we collect

Account details you provide (name, email, and a password hash or the identifier from Google/GitHub if you sign in with them); the website URLs you add; monitoring results and audit findings for those sites; screenshots and generated reports; content you draft or publish through Discover; and billing records. Payment card details are handled entirely by Stripe and are never seen or stored by Zenith.

How we use your data

To run monitoring and audits on the sites you add, send transactional email (verification, alerts, receipts), operate billing, secure the service, and provide support. We do not sell your data or use it for advertising.

Monitor prompts

Monitor's “Fix with Claude” prompts are generated deterministically on Zenith's servers from audit findings; the prompt-generation step does not call an AI provider. You choose whether to paste a prompt into your own AI tool; that action happens outside Zenith.

Marketing email

Marketing email (product updates, monitoring tips, growth insights, security updates) is entirely separate from your account email and requires your explicit double opt-in: creating an account never subscribes you. When you subscribe — from the site footer or Settings — we send a confirmation link before anything else arrives, and store your email address, subscription status, the time and source of your consent, when you confirmed, your chosen topics, and if applicable when and why you unsubscribed. Every marketing email carries a one-click unsubscribe link; unsubscribing takes effect immediately and doesn't affect account email about your sites or billing.

AI processing (Discover)

Discover's AI-assisted drafting is off by default and per account. If you enable it in Settings, Zenith uses the selected provider shown there: Anthropic when an Anthropic API key is configured, otherwise Cloudflare Workers AI when its binding is available. Shared site context contains the profile's name, description, category, audience, brand tone and up to 12 primary topics; the target keyword; and up to 8 recently crawled public pages. For each page, the prompt includes its URL and one label: its title when present, otherwise its H1, otherwise its URL. It can also include the page's meta-description excerpt (up to 400 characters). The stored secondary-topic list and page headings are not included as shared context.

Depending on the action, a prompt can also include a current brief's angle and outline; a draft's title, outline and existing body (up to 4,000 characters); a full-article request's existing excerpt, metadata, secondary keywords, search intent and CTA; or a metadata request's page title and content excerpt (up to 1,000 characters). The application does not add account credentials, billing details or private reports to these prompts. Generated content is returned as a proposal for review; applying it and publishing are separate actions. You can turn AI processing off in Settings for later requests.

For current provider terms, see Cloudflare Workers AI data usage and Anthropic API data retention. Cloudflare describes standard inference as not retaining inputs or outputs after processing and says it does not use Workers AI customer content to train models, with stated exceptions for storage features and legal or contractual requirements. Anthropic describes standard API input/output deletion within 30 days, subject to stated exceptions, and says retained data is not used for model training without express permission. Zenith has not verified any account-specific provider agreement or retention setting.

Search Console (optional)

If you connect Google Search Console (Settings, Integrations), we store the OAuth tokens (encrypted), your Google account email for display, and read-only search performance data — clicks, impressions, position and the queries and pages behind them — for the property you choose for each site. We request the read-only webmasters.readonly scope only, through a connection separate from signing in with Google. Detaching a site's property (the site's Settings) stops syncing it and deletes that site's imported data. Disconnecting Search Console (Settings, Integrations) deletes our tokens and every site's imported data, and revokes Zenith's access at Google unless Google Analytics is still connected (Google can treat both connections as one grant, so it's revoked when you disconnect the second).

Bing Webmaster Tools (optional)

If you connect Bing Webmaster Tools (Settings, Integrations), we store the API key you paste (encrypted) and import, for each Site whose address matches a site in your Bing account, daily clicks and impressions and the search queries behind them. Bing's API keys give access to your whole Bing Webmaster account and can't be limited to reading; we only ever call Bing's read methods, and you can revoke the key in Bing at any time. Disconnecting deletes our copy of the key and every imported Bing row.

Your Google Analytics data (optional)

If you connect Google Analytics (Settings, Integrations), we store the OAuth tokens (encrypted) and your Google account email for display, and, for the GA4 property you choose for each Site, import daily sessions, page views and active users and per-page totals. We request the read-only analytics.readonly scope only, separately from Search Console and from signing in with Google. This is your own analytics data about your site; it's never used for Zenith's own analytics. Detaching a property deletes that Site's imported data. Disconnecting deletes our tokens and every Site's imported data, and revokes Zenith's access at Google unless Search Console is still connected.

GitHub source repository connection (optional)

If you connect the GitHub source App to a verified site, Zenith stores GitHub user and installation account IDs, login names, account type and connection status; encrypted user access and refresh tokens; and the selected repository's ID, owner, name, default branch, access status and last-check time. Authorization records hold a state digest, expiry and use status. Zenith rechecks whether the authorized GitHub user can access an installation and its repositories, and records signed webhook delivery IDs, event names and received times for installation/repository lifecycle updates. The source connection does not retrieve or change repository contents. Removing it from Zenith deletes its local connection and site mappings; it does not uninstall the App or revoke the authorization in GitHub. To remove GitHub's access, uninstall or revoke it in GitHub as well.

Keyword data

To estimate search volume and competition for keywords related to your site, Zenith sends the keyword text you request estimates for to DataForSEO, our keyword-data subprocessor, and caches the response.

Publishing destinations and alert channels

If you connect a publishing destination (WordPress, Ghost, Webflow, Shopify, GitHub or a webhook), any credential you give us (an API key, access token, application password or webhook secret) is encrypted at rest and used only to publish the content you choose to that destination. Alert channel URLs (Slack, Discord, Microsoft Teams or a webhook) are encrypted the same way and used only to send your uptime alerts: the site's name and address, whether it went down or recovered, the status code or error, and a link back to Zenith.

Zenith Discover pages

If you publish to Zenith Discover, the resulting page is public content you chose to publish and are responsible for. Published pages are served from discover.zenith.me.uk, a separate address from your Zenith account. We may remove content that breaches the Acceptable Use Policy and suspend publishing for repeated breaches.

What we don't retain

We do not keep long-term full copies of your website's HTML. Audits store the findings, limited evidence samples, and screenshots needed for your report.

Analytics

We keep a small amount of privacy-friendly first-party usage data (which pages and actions are used) that does not rely on tracking cookies; unless you accept analytics, the random id that groups those events lives only in memory for the current page and is never stored on your device. Separately, with your consent, we use Google Analytics 4 to understand site usage; it loads only after you accept in the cookie banner, IP addresses are anonymised, and you can decline, or withdraw consent at any time from Cookie settings in the site footer.

Lawful bases

We process account, monitoring, audit and billing data to perform our contract with you; transactional email and security processing on the basis of legitimate interests; and we rely on your consent for any optional communications.

Processors

We use a small set of providers to run the service. See the Subprocessors page for the current list and what each one processes.

Retention

Account and site data are kept while your account is active. Discover content items, page analyses and search-performance metrics are kept for as long as the site they belong to exists, and are deleted with the site or the account. When you delete your account, your sites, screenshots, reports, content and monitoring data are removed, monitoring stops, and we revoke Zenith's access to any Google account you connected. Billing records are retained only as long as required for tax and accounting. Operational logs are short-lived.

Your rights

You can export your data — including your email preferences and AI-processing setting — and permanently delete your account at any time from Settings; deletion also removes your marketing-email subscription record. Depending on your location you may also have rights to access, correct, restrict or object to processing, and to lodge a complaint with your local data-protection authority.

International processing

Our providers may process data in the UK, EU and US under appropriate safeguards. See Subprocessors for detail.

Contact

Questions or requests: support@marcelipawlinski.co.uk.