Flareboard
HomeFeaturesComparePricingBlogDocs
Sign inStart free

Privacy Policy

English简体中文

Last updated: October 8, 2026

Contents

  1. Who we are
  2. Two kinds of data, two roles
  3. Customer data we collect
  4. Analytics on our own website
  5. Analytics data we process for customers
  6. How we use customer data
  7. AI assistant and MCP access
  8. Legal bases (EEA and UK)
  9. Cookies and browser storage
  10. Service providers and sharing
  11. International transfers
  12. Retention and deletion
  13. Your rights
  14. Security
  15. Children
  16. Changes to this policy
  17. Contact

This policy explains what personal data Flareboard collects, why, how long we keep it and the choices you have. We built Flareboard to measure websites without cookies and without storing IP addresses, and we try to collect as little as the service needs.

1. Who we are

Flareboard (“Flareboard”, “we”, “us”) is a product analytics service operated by an independent developer. This policy covers the hosted service at flareboard.dev, including our website, the dashboard, the API (api.flareboard.dev) and the tracking endpoint (t.flareboard.dev). You can reach us about anything in this policy at support@flareboard.dev.

Flareboard is also published as source code that anyone can run on their own infrastructure. If you use a self-hosted Flareboard, the person or organization running it decides how data is handled; we have no access to that data and this policy does not apply to it.

2. Two kinds of data, two roles

We handle personal data in two different capacities:

  • Customer data. Information about people who visit our website or create a Flareboard account (“customers”). For this data we decide why and how it is processed, so we act as the controller.
  • Analytics data. Information our tracking script and APIs collect about visitors to our customers’ websites and apps (“visitors”). Our customers decide to install Flareboard and what to collect; we process this data only to provide the service to them. For analytics data the customer is the controller and we are a processor (or service provider) acting on the customer’s instructions, as set out in our Terms of Service.

If you are a visitor to a website that uses Flareboard and want to exercise your privacy rights, please contact that website’s owner first. We will help them respond.

3. Customer data we collect

Account information

  • Email address (also used as your username), optional display name and logo.
  • Your password, stored only as a salted bcrypt hash. We never store or see your password in plain text.
  • If you sign in with Google or GitHub (when offered), the provider’s account identifier and the email or username the provider returns. We do not store the provider’s access tokens. We connect a provider account to an existing Flareboard account automatically only when the provider and Flareboard have both verified the same email address.
  • If you turn on two-factor authentication, the secret your authenticator app uses, stored encrypted, and your one-time recovery codes, stored only as one-way hashes. Turning two-factor authentication off deletes both.
  • Email verification status, role, team memberships and account timestamps.
  • Personal API keys you create: their name, scopes, the first characters of the key and when it was created and last used. The full key is shown to you once and stored only as a one-way (SHA-256) hash. Revoking a key deletes it; the rest are erased with your account.

Billing information

Payments are handled by Stripe. Card details are entered on Stripe’s pages and never reach our servers. We receive and store your Stripe customer and subscription identifiers, plan, subscription status and billing period, and we send Stripe your email address and account identifier to create the subscription. We also count the events, session recordings and OpenTelemetry log records and spans your websites send each month, to apply your plan’s allowances and to email you about your usage. When collection of a kind of data has stopped for the month, what your websites send of it is discarded, not stored.

Content you create

Websites, dashboards (“boards”) including their filters and layout, reports, insights, notebooks, segments, cohorts, goals, feature flags, experiments, surveys, workflows, alert rules, saved queries, annotations, uploaded source maps and similar configuration you create in the dashboard.

When you set up an insight alert or an email subscription to a board or insight, we store the email addresses or webhook URL you enter, the schedule, and a history of alert checks (the checked value and whether a notification was sent). Subscription emails contain aggregate numbers only. Alerts and subscriptions are deleted with their board, insight or website, and subscriptions you created are deleted with your account.

Activity and security records

  • An audit log of administrative and security actions in your account, recording who did what and when: for example creating or deleting a website, exporting data, creating share links or API keys, changing team members or roles, sign-ins and failed sign-in attempts (with the sign-in method and a coarse device description such as “Chrome on macOS”), password changes, two-factor authentication changes and signing out sessions. For feature flags it also keeps the flag’s settings before and after each change. It does not record IP addresses. You can see your own account’s activity in the dashboard, and owners and managers of a team can see the team’s activity, including which member made each change. Entries about a website are erased with that website.
  • Your signed-in sessions: for each one, a coarse device description derived from the browser’s user agent (such as “Chrome on macOS”; the user agent itself is not stored), how you signed in, and when the session started and was last active. No IP address is stored. You can review and sign out sessions in the dashboard. Opening the demo (“Explore the demo”) without an account starts the same kind of session for a shared, read-only demo account; it holds nothing else about you, cannot change anything and ends after four hours.
  • Technical logs generated by our hosting provider when our servers handle requests, such as request paths, status codes and error messages. These may contain IP addresses and are kept for a short period (see Retention).
  • IP addresses are used briefly for rate limiting and abuse prevention (for example, limiting sign-in attempts and temporarily locking sign-in after repeated failures). We do not store them in our database. Sign-in lockout counters are keyed by a one-way hash of the address or account name, never the value itself, and are deleted within two hours of the last failed attempt.

Communications

Messages you send us, for example support emails, and our replies.

4. Analytics on our own website

We measure how people use flareboard.dev so we can improve it. We use two tools, neither of which sets cookies:

  • Flareboard itself. Our own tracking script records the same kinds of data described in Analytics data below (pages viewed, referrer, browser, device type, approximate location), plus a small number of product events such as starting a sign-up, verifying an email address or a successful sign-in. These events carry a label and an error code at most, never your email address or password.
  • Cloudflare Web Analytics. Cloudflare measures page load performance and aggregate visits. See Cloudflare Web Analytics.

5. Analytics data we process for customers

When a customer adds the Flareboard script or sends data through our APIs, we process the following on their behalf. Customers choose which features to enable, so not every site collects everything below.

Collected automatically by the tracking script

  • The page address (path, query string and fragment), page title and hostname, plus the referring page and any campaign parameters (such as utm_source) or ad click identifiers (such as gclid) present in the address.
  • Browser, operating system and device type (derived from the user agent; the full user agent string is not stored), screen size and browser language.
  • Approximate location (country, region and city), derived by our hosting provider from the IP address at the time of the request.
  • Page performance metrics (Core Web Vitals), when enabled.
  • Aggregated click positions for heatmaps, when enabled. Heatmap data is not linked to a visitor.
  • Interactions (autocapture), when the customer leaves it on: clicks on links and buttons, form submissions and changes to form fields. For each, the script records the kind of element, its identifier, name and CSS classes, the link address, a short description of its position in the page and up to 255 characters of its visible text. It never records what is typed or selected in a form field, never reads text from form fields, password fields or elements that look sensitive (for example payment or one-time-code fields), drops text that looks like a card number, identity number or email address, and records nothing inside parts of a page the customer marks as excluded.
  • Page leave, together with autocapture: how long a page was visible and how far down it was scrolled.

How visitors are counted

IP addresses are never stored. To count unique visitors without cookies, we compute a one-way hash of the IP address, user agent, website identifier and a secret value. The secret value rotates at the start of each calendar month, so the same browser on the same network is counted as one visitor within a month but cannot be linked across months. The IP address itself is discarded once the request is handled.

Remembering visitors (off by default). A customer can turn on “Remember visitors across sessions” for a website. The script then creates a random identifier, keeps it in the browser’s localStorage and sends it with each event, and that identifier is used instead of the hash above to count the visitor. The same browser is then recognized across networks and months, until the visitor clears the site’s data or the website resets it (for example at sign-out). The identifier is random: it is not derived from the IP address or the device. IP addresses are still never stored. Because this stores an identifier on the visitor’s device, customers who turn it on are responsible for obtaining any consent the law where their visitors are requires.

Customers can also send data with PostHog’s open-source SDKs (such as posthog-js or posthog-node) pointed at Flareboard. We process it as described on this page: IP address properties and raw user agent strings included in such data are discarded, not stored. Those SDKs may use cookies or localStorage on the customer’s site according to how the customer configures them.

The tracking script sets no cookies. It keeps short-lived session information in the browser’s sessionStorage, which is cleared when the tab is closed. It uses localStorage only when a customer identifies a signed-in user or shows a survey (to remember that identifier or survey state), when a visitor opts out (to remember that choice), and, on websites that remember visitors, for the random visitor identifier and the event properties the website asks the script to attach to every event.

Data customers choose to send

  • Custom events and properties, including user identifiers and profile properties sent with identify(), group membership and revenue amounts. Customers control this content and may include personal data such as an email address. When a visitor opts in or out of an early access feature, that choice is stored as a property of their profile.
  • Session replay (off by default; paid plans only). When a customer turns it on, the recorder script records the structure of the page and the visitor’s interactions (clicks, scrolling, mouse movement and page changes) so the session can be replayed. By default, everything typed into form fields is masked. Password, payment-card and one-time-code fields and fields that look sensitive stay masked even if a customer turns input masking off. Customers can mask all text on the page, mask or block further parts of the page (blocked elements are replaced by an empty box of the same size, and elements marked data-fb-no-capture or ph-no-capture are never recorded), record only a sample of visits and skip visits shorter than a minimum duration. Text shown on the page is recorded unless masked, so customers must configure replay not to capture sensitive information. Nothing is recorded for visitors who opted out, or who send Do Not Track or Global Privacy Control where the website honors those signals.
  • Console and network activity in replays (off by default; each is a separate customer setting). With console capture, the level and text (up to 1,000 characters) of messages the page writes to the browser console, and of uncaught errors, are stored with the recording. Email addresses, card-like numbers, access tokens, values labeled as passwords or secrets and the query part of URLs are removed in the browser first. With network capture, each request the page makes with fetch or XMLHttpRequest is stored as its method, address without the query string, status code, duration and size. Request and response headers and bodies are never recorded, and our servers discard any other field before storing a recording.
  • Errors and logs: error messages, stack traces, release and environment names, and log messages with the data customers attach.
  • OpenTelemetry logs and traces that customers send from their own servers and apps: timestamps, severity, the log message, service name, version and environment, trace and span ids, span names and durations, and the attributes the customer's software attaches, which can include identifiers such as a session or user id. A session id links a log line or trace to that session and its replay. Flareboard does not add IP addresses to these records; attribute counts and sizes are capped.
  • Survey responses, including free-text answers visitors choose to submit. If a visitor closes a multi-question survey part-way, the answers already given are kept as a partial response. Customers can also share a hosted survey link that works without the tracking script. Answers sent there are stored the same way, together with any identifier the customer adds to the link, and the page remembers in the browser's localStorage that the survey was answered.
  • AI usage data (LLM analytics): model, provider, token counts, cost, latency, status, errors and the structure of each trace (trace, span and parent ids and names). Prompts and responses, and the input and output of traced steps, are stored only when a customer sends them (for example through the PostHog LLM wrappers); each is cut to 32 KB. A customer can turn off content storage for a website, in which case only the metadata above is kept. Stored content follows the website’s retention setting.
  • Data warehouse imports a customer sets up: rows fetched from HTTP JSON or CSV addresses the customer configures, and, when a customer connects their own Stripe account, that account’s customers (email, name and metadata), subscriptions, invoices and invoice line items, charges and refunds. We import this on the customer’s behalf with a restricted, read-only API key the customer creates. Before storing Stripe records we drop payment method and card details, billing and shipping addresses, phone numbers, tax IDs and links to hosted invoices. A Stripe customer is linked to a person in the customer’s analytics when the Stripe metadata carries that person’s distinct_id or the email address matches a profile, so revenue can be attributed to how that person first arrived.

The tracking script ignores most automated traffic (bots and crawlers). Customers can set a website to honor Do Not Track and Global Privacy Control browser signals, in which case the script sends nothing from browsers that send either signal. Websites can also offer visitors an opt-out; the choice is remembered in the browser and neither the tracking script nor the session recorder then sends anything.

We do not use analytics data for our own purposes, do not combine it across customers, and never sell it or use it for advertising.

6. How we use customer data

  • To create and secure your account, sign you in and provide the service you signed up for.
  • To process payments, apply plan limits and manage your subscription.
  • To send service emails: email verification, password resets, usage notices to the account owner (when usage reaches 80% and 100% of a monthly allowance and when collection stops), and alerts, reports or workflow emails you configure. We do not send marketing email without your consent, and you can opt out at any time.
  • To answer support requests.
  • To prevent abuse, fraud and security incidents, and to debug problems.
  • To understand how the product is used, in aggregate, so we can improve it.
  • To comply with legal obligations and enforce our Terms of Service.

We do not sell personal data and do not share it with advertisers.

7. AI assistant and MCP access

Where the operator has enabled it, the dashboard offers “Ask Flareboard”, an assistant that answers questions about a website’s analytics. It is off unless you use it, and it only runs when you send it a question.

  • What is sent to DeepSeek. To answer, we send Hangzhou DeepSeek Artificial Intelligence Co., Ltd. (the provider of the DeepSeek models) your question, up to the last 12 messages of the same conversation, the website’s name, domain and timezone, and the results of the queries the assistant runs for you. Those results are limited in size (at most 100 rows per query and a capped number of characters) and can include analytics data such as event names, page URLs, property values and, when a question needs it, the identifiers, email addresses or names of people recorded by the website. Only read-only queries run on behalf of the assistant; it cannot change your data.
  • How DeepSeek uses it. DeepSeek processes this data as our subprocessor to generate the answer and stores it on servers in the People’s Republic of China. Under its own terms and privacy policy, DeepSeek may retain inputs and outputs and may use them to improve its models. Do not use the assistant for questions whose results you are not willing to share with DeepSeek; if you or your visitors are subject to rules that restrict transfers to China, leave the assistant unused.
  • What we store. Your questions and the assistant’s answers (with the charts and tables they show) are stored in your account, visible only to you, for that website. We also count how many questions each account asks per day and the number of tokens used, to apply usage limits. We do not write questions or answers to our logs.
  • Retention. You can delete a conversation at any time from the assistant panel. Conversations without activity for 90 days are deleted automatically, and all of them are erased with the website or your account.
  • MCP access. You can connect AI tools (such as Claude Desktop, Claude Code or Cursor) to Flareboard through our Model Context Protocol endpoint with a personal API key. The data those tools read is sent to the tool and AI provider you chose, under their terms, not ours. Changes made through MCP (annotations, feature flag toggles) are recorded in the audit log.

8. Legal bases (EEA and UK)

If you are in the European Economic Area or the United Kingdom, we rely on these legal bases for customer data:

  • Contract: to provide the service, manage your account and process payments.
  • Legitimate interests: to secure and improve the service, prevent abuse, measure use of our website and communicate with you about your account. We balance these interests against your rights, and you can object at any time.
  • Legal obligation: to keep records required by tax and accounting law.
  • Consent: where the law requires it, for example for optional marketing emails. You can withdraw consent at any time.

For analytics data, the customer who installed Flareboard determines the legal basis.

9. Cookies and browser storage

NameWherePurposeDuration
flareboard_sessionDashboard (cookie)Keeps you signed in (or in the read-only demo). Strictly necessary; HttpOnly and Secure.7 days (4 hours for the demo)
__cf_bm, cf_clearanceAll our domains (cookie)Set by Cloudflare to detect bots and protect the service. Strictly necessary.Up to 30 minutes / as set by Cloudflare
Theme, language and interface preferencesDashboard (localStorage)Remembers your display choices.Until cleared
flareboard.*Websites using Flareboard (sessionStorage, localStorage)Session information for the tracking script, as described in Analytics data. No cookies.Until the tab closes; identify, survey and opt-out state, and (when the website remembers visitors) the random visitor identifier, until cleared

We do not use advertising or cross-site tracking cookies.

10. Service providers and sharing

We share personal data only with the service providers we need to run Flareboard:

ProviderPurposeData
Cloudflare, Inc.Hosting, databases, file storage, queues, content delivery, security, email delivery, logs and web analyticsAll customer and analytics data
Stripe, Inc.Payments and subscription billingEmail, account identifier, payment details (entered directly with Stripe)
Google LLC, GitHub, Inc.Optional sign-in, only if you choose itAccount identifier, email or username, and whether the provider verified the email
jsDelivr, unpkg, CARTO, Simple IconsFonts, map data and icons loaded by your browser when you use the dashboardYour IP address and browser details, as part of the request
Hangzhou DeepSeek Artificial Intelligence Co., Ltd. (China)Answers from the “Ask Flareboard” assistant, only when you use it (see AI assistant)Your questions, recent conversation messages and capped query results about the website

We may also disclose personal data:

  • At your direction, for example to webhook URLs, email addresses or data sources you configure, or through public share links you create for a website, board or insight. Anyone with a share link can see the aggregate statistics it exposes until the link expires or you revoke it. Anyone with a replay link can watch that one session recording, including any console and network activity it holds, until the link expires or you revoke it. Workflows send what their templates include to the webhook, email and Slack destinations the customer configures: by default the triggering event with its properties, page URL and distinct ID, and the stored properties of that person. Webhook deliveries are signed so the receiver can verify they came from Flareboard. When you connect a Stripe account as a data source, we call Stripe’s API with your key to read your data; Stripe acts as your provider there, not ours.
  • For legal reasons, if required by law or a valid legal request, or to protect the rights, property or safety of our users, the public or us.
  • In a business transfer, if Flareboard is sold or transferred, in which case this policy will continue to apply to your data.

11. International transfers

Flareboard runs on Cloudflare’s global network, so data may be processed in data centers in many countries, including outside your own. Where the law requires it, we rely on appropriate safeguards for these transfers, such as the European Commission’s Standard Contractual Clauses included in our providers’ data processing terms. If you use the optional “Ask Flareboard” assistant, the data it sends is transferred to DeepSeek in the People’s Republic of China (see AI assistant).

12. Retention and deletion

  • Account data is kept while your account is open. You can delete your account at any time from the account menu in the dashboard; it is closed immediately, and your account and content are permanently erased within 30 days.
  • Security records. Sign-in sessions are deleted when you sign out or revoke them, or when they expire (7 days after they were last renewed; demo sessions after 4 hours). Records of sign-ins and failed sign-in attempts are deleted after 180 days. Other audit log entries are kept while your account (or, for entries about a website, that website) exists.
  • Billing records are kept for as long as tax and accounting law requires, generally seven years.
  • Analytics data (raw events and session replay recordings) is kept for each website's retention period and then deleted automatically, unless the customer deletes it sooner. The customer can set a period for each website, up to the maximum of their plan: 1 year on the Free plan, 2 years on the Cloud plan and 3 years on the Business plan. When no period is set, the plan maximum applies. If an account moves to a plan with a shorter maximum, raw data older than that maximum is deleted. Session replay recordings are deleted together with the raw events they belong to. Aggregated statistics derived from raw events may be kept for the life of the website.
  • Imported warehouse data (HTTP and Stripe imports) is kept until the customer deletes the data source or the website; the per-website retention period does not apply to it. Deleting a data source erases what it imported and its stored API key right away. Replacing a Stripe API key erases the data imported with the old key.
  • OpenTelemetry logs and traces are deleted automatically after 30 days, or sooner when the website's retention period is shorter. Records older than 30 days are not accepted.
  • Deleting a website in the dashboard removes it from your account immediately, and all of its stored data, including session replay recordings, imported warehouse data and stored API keys, is permanently erased within 30 days.
  • Workflow execution logs (which event triggered a run, each delivery attempt, the destination's response code and the first 1,000 characters of its response) are deleted after 90 days.
  • Assistant conversations are kept until you delete them, and are deleted automatically after 90 days without activity or with the website or your account.
  • Technical logs kept by our hosting provider are deleted automatically, typically within 7 days.
  • Backups. Deleted data may remain in database backups for up to 30 days before it is overwritten.

13. Your rights

Depending on where you live, you may have the right to access, correct, delete or export your personal data, to object to or restrict certain processing, and to withdraw consent. You can view and edit most account information in the dashboard, delete your account from the account menu, and paid plans can export analytics data as CSV. For anything else, email support@flareboard.dev from the address on your account. We will respond within 30 days.

If you are unhappy with how we handle your data, please tell us first. You also have the right to complain to your local data protection authority, such as a supervisory authority in the EEA, the UK Information Commissioner’s Office, or the Office of the Privacy Commissioner for Personal Data in Hong Kong.

Visitors to customers’ websites: we do not hold a direct relationship with you and usually cannot identify you. Please contact the owner of the website you visited; we will assist them in responding to your request.

14. Security

We protect data with encryption in transit (HTTPS), the encryption at rest provided by our hosting provider, hashed passwords and API keys, encrypted and revocable session tokens, optional two-factor authentication (which team owners can require), role-based access controls, rate limiting and temporary lockout after repeated failed sign-ins. No system is perfectly secure, however, and we cannot guarantee absolute security. If a breach affects your personal data, we will notify you and the relevant authorities as required by law.

15. Children

Flareboard is a business tool and is not directed at children. You must be at least 16 to create an account. We do not knowingly collect personal data from children; if you believe a child has given us personal data, contact us and we will delete it. Customers must not use Flareboard to collect data from children in violation of applicable law.

16. Changes to this policy

We may update this policy as the service changes. We will change the “Last updated” date above, and if a change is significant we will notify customers by email or in the dashboard before it takes effect.

17. Contact

Questions or requests about privacy: support@flareboard.dev.

Flareboard

Product analytics on Cloudflare. Use the hosted cloud or deploy it on your own account.

Product

FeaturesComparePricingLive demoMac app

Resources

DocsBlogSelf-hostGitHub

Account

Sign inStart free
© 2026 Flareboard
Terms of ServicePrivacy Policy