What a technical audit stores
When you submit a public URL and brand, we store the normalized URL without its fragment, the brand name, technical observations, score, report and timestamps. An optional email unlock stores your lowercase email address and consent time against the report. No email is sent by this unlock, and it does not subscribe you to the waitlist.
The audit fetches the submitted page and supporting robots.txt, sitemap and llms.txt files. It uses their response bodies for raw-HTML technical checks; it does not retain page bodies. Do not submit confidential URLs or personal information in a URL or brand field. We do not fetch AI product interfaces or Google search pages, execute page JavaScript, train models or sell submitted data. Logs do not contain the URL, brand, email, report or other user content. There is no upload interface.
Private links and report deletion
Each report has a random private link. It is excluded from indexing and the sitemap, but anyone who obtains the link can view the available report. Keep it private if you do not want to share the result.
The creation response gives your browser a deletion credential once. It stays in the current document’s memory, is sent only in request bodies and is never included in the private link. The database retains a derived value instead of the credential itself. Use “Delete this report,” then confirm, to remove the report and all linked emails. Refreshing, closing the tab or opening the link in a new tab loses the credential; a private link does not recover it.
Without that credential, contact contact@citevisor.com for a deletion request. We may request enough information to verify your connection to the report before deleting it. Otherwise, the report expires under the retention schedule below.
Why we keep it and for how long
We use audit data to provide the technical checklist and repair instructions. Unlocked reports and their linked emails are kept for up to 30 days after unlock; reports not unlocked are kept for up to 24 hours after creation. A daily cleanup removes expired records, so removal occurs at the next scheduled cleanup. Expired reports are unavailable immediately. These are the default limits; any operator configuration change must keep this notice accurate.
We reuse a normalized URL’s recent technical observations for 60 minutes by default to reduce repeat website requests. Aggregate daily audit, unlock and cache-hit counts measure service use without recording raw IP addresses or page content.
For visitor limits, we use the client IP transiently to derive a short-lived rate-limit key. The database stores the derived key and submission timestamps, never the raw IP. These records expire after 23 hours and are removed by hourly cleanup or subsequent audit submissions, keeping normal retention within 24 hours. An interruption to scheduled cleanup can delay physical deletion and requires operator follow-up.
What the waitlist stores
The separate waitlist collects your normalized email address, role, plan interest, optional website origin, consent and signup timestamps, campaign source, medium, campaign, term and content values, and a referring domain when available. The optional website is validated and stored as an origin only; submitting it to the waitlist does not fetch it.
We use waitlist and update signup details to understand interest and send product updates, including launch news. Joining is not a purchase. Unconverted entries expire after 12 months and are removed by the daily cleanup. Withdraw consent or request deletion by emailing contact@citevisor.com from the signup address. We do not store full referrer URLs or raw IP addresses. Please do not put sensitive information in campaign parameters.
What an account stores
You sign in with a one-time link we email to you or with your Google account; no password is stored. A sign-in link works once, for 15 minutes, and only a derived value of it is kept. If you continue with Google, Google tells us your email address and whether it has verified it; we use that only to sign you in to the account with that address, ask for no other Google data and store nothing else from Google. An account keeps your email address, its creation and last sign-in times, your report email preference and a derived value of each active session; sessions last 30 days, and signing out ends them. A second cookie, holding no credential, only tells this site’s pages to show your dashboard link instead of Sign in. Plans you buy or our team sets up and monthly usage counts are kept with the account.
Each project keeps the website, brand, other brand names, competitors, prompts, engines and language you enter, and every run’s results: the technical check of your page, the answers and citations each AI engine returned, whether and where your brand was named and a summary of the run. Do not put sensitive or personal information into prompts or project fields.
How long account data is kept
Answer text, full citation lists and usage details are cleared 180 days after each check; mention results, positions and run summaries, including which of your pages were cited and which other domains answers relied on, stay for your trend. Projects and their runs stay until you delete the project or the account. Email jobs are deleted 30 days after they finish. Accounts without a current plan that have not signed in for 18 months are deleted by the daily cleanup.
Emails we send to account holders
We send the sign-in links you request and, unless you turn them off, a short report after each finished weekly run. Every report email has a one-click unsubscribe link, and your account has the same switch. To keep that choice working, we keep a derived value of an unsubscribed address for each kind of email, even after the account is deleted. Cloudflare Email Sending delivers these emails.
AI engines
Each run sends your prompts, and nothing else from your account, to OpenAI, Perplexity, Google (Gemini) and Anthropic (Claude) through their APIs with web search, and stores the answers they return. Requests may pass through OpenRouter, an API gateway that forwards each prompt to that engine’s own provider; we ask it to use only providers that do not train on prompts. Your email address and account details are never sent to them; the API terms of the engines and of OpenRouter govern their processing. Answers through an API can differ from what their consumer apps show.
Agency white label and share links
An Agency account can save a display name and accent colour for its reports and create read-only links to a project’s report for its clients. For each link we keep only a derived value of it, its label and its dates. A shared report shows the white label name, the project’s brand and website, run summaries (including competitor names, cited pages and the questions whose results changed), the trend and the technical check; it never shows the account email, other projects or answer text. Anyone who holds a link can open it until it expires (after at most 365 days), it is revoked, the project or account is deleted, or the account no longer has the Agency plan. Expired links are deleted by the daily cleanup. Opening a shared report counts toward a per-visitor limit, kept as a derived value of the IP address for up to two hours, and the page loads no analytics tags.
Your account data, your control
Download everything your account holds from your account page as one file. Delete a project to remove its runs and answers, or delete your account to remove every project, run, answer, plan record and pending email. Deletion cannot be undone. Unsubscribe records stay as described above.
What a purchase stores
Purchases are made from your account. When you start one, we store the plan or report you chose, your account, the website you enter for a repair report as an origin only (no path or query) and the time. When the payment is confirmed we store payment and subscription identifiers, the amount, the subscription status and paid period, refunds and report delivery. Payments run through billing-hub, a payment service we operate on Cloudflare: it receives your account identifier and email address to open Waffo’s checkout, and keeps the payment ledger. Waffo, the merchant of record, collects your billing details and payment method on its checkout page under its own privacy policy. We never receive card numbers.
Unfinished purchase attempts are deleted after 7 days by the daily cleanup. Payment records are kept for as long as we need them to run the subscription, deliver the report, handle refunds and meet legal and tax obligations, including after you delete your account (without your email address in this site’s records). Email contact@citevisor.com to request access or deletion; we may keep what the law requires.
Service providers
Cloudflare Workers hosts the service, D1 stores reports, accounts and monitoring data, waitlist and order records, and Turnstile verifies submissions to reduce automated abuse. Cloudflare Email Sending delivers sign-in and report emails. Google provides the optional Google sign-in under its own privacy policy. billing-hub, our payment service on Cloudflare, records payments, and Waffo processes them as merchant of record. For real website audits, Cloudflare’s public DNS-over-HTTPS resolver receives hostnames to check public-network destinations before fetching. It does not receive your email, brand or report. MOCK mode instead uses synthetic fixtures and makes no website or DNS request.
Cloudflare Turnstile processes the technical information needed for verification under its own privacy practices. Google Analytics 4 measures page views and on-site interactions such as scrolling, outbound link clicks and the events described below; if Google Ads measurement is configured and you accept optional cookies, Google also processes those events for advertising measurement. Google tags never run on private report pages, account, sign-in or unsubscribe pages: opening a report switches them off for the rest of that page session, so report links are not sent to Google. Providers may process information outside your country. Final operator identity, any required DPA and regional transfer terms remain to be confirmed before a public launch.
Cookies and your choice
Your accept or reject choice is stored in this browser as citevisor_ad_consent, separately from report or waitlist consent. Google Analytics is enabled by default. In the EEA, UK and Switzerland, Google receives measurement signals without analytics cookies until you allow them. Optional Google Ads measurement requires acceptance. Analytics sets first-party cookies (_ga and _ga_*) to tell visits apart; analytics event data is kept for two months and user-level analytics data for up to 14 months. Google Analytics 4 does not log or store IP addresses. Besides page views and standard interaction events, we send only waitlist_submit after successful signup, presale_click when a presale button is clicked, audit_complete when a free audit result appears, report_unlock when a report is unlocked, and buy_click when a purchase button is clicked. buy_click carries only the product code (for example repair_report); no event contains emails, audited websites, brands or form answers.
Rejecting optional cookies leaves the audit and waitlist available and loads no Google tag. Use Cookie settings to choose again. Turnstile protects submissions whether or not optional measurement is accepted. No other analytics tracker is installed.
Checkout and browser tool
Purchase buttons open Waffo’s hosted checkout page; this site loads no payment script. The crawler rule generator runs in your browser without fetching your site or changing its robots.txt.
Local full mode is separate
The Phase 1 full build stores public website URLs, brands, prompts, answers, citations and email unlock details in the operator’s local SQLite database and monitoring reports. Its deletion control removes associated records, and refreshing loses its in-memory credential. This full build and its model adapters are excluded from the landing Worker.