Skip to content
Legal

Privacy Policy

Effective date: January 1, 2026 · Last updated: November 22, 2026

Summary (Plain Language)

PerfBee is a B2B web performance analytics platform. We collect only what we need to operate the service. We do not sell your data and we do not use it for advertising. Our Real User Monitoring (RUM) script does not collect names, email addresses or IP addresses from your visitors. Analytics only starts after you allow it in the cookie banner.

1. Who We Are & Our Role

Techpiks Bilgi Teknolojileri Ticaret Limited Şirketi (Techpiks Information Technologies Ltd.), operating as Techpiks Digital Solutions (“PerfBee”, “we”, “us”, or “our”) operates the web performance analytics platform PerfBee, available at perfbee.com and dashboard.perfbee.com.

Address: Ritim İstanbul, Cevizli, Zuhal Cd. A5 Blok No: 46E/179, 34846 Maltepe/İstanbul, Türkiye

  • Data Controller: For account data and platform usage data we collect directly from you (the customer), and for visitors of our own websites.
  • Data Processor: For RUM visitor data collected on your website on your behalf. In this capacity, you (the customer) are the data controller and are responsible for ensuring a valid legal basis under GDPR for collecting visitor performance metrics. We act in the same capacity for the visitor data in a server log file that you upload to Bot Tracker (see 2.12).

Contact: support@perfbee.com. We have not appointed a Data Protection Officer; privacy requests go to this address.

2. Data We Collect

2.1 Account Data

When you create an account, we collect your name, email address and password (stored hashed) and, if you add them, your phone number and company name. If you sign in with Google, GitHub or Shopify, we receive your name and email address from that provider. Payments are handled by our payment provider: we receive your billing details and subscription status, and we never store card numbers on our servers.

2.2 Usage Data

We collect information about how you interact with the platform: pages visited, features used, audit results, credit consumption, and support requests. If you ask for an export that is too large to build while you wait, we build it as a file on our servers and keep that file for up to 72 hours so that you can download it. Then we delete it.

2.3 Technical Data

IP address, browser type and language, device type, operating system, and session identifiers, collected in server logs and by our security tools.

2.4 Free Tools and Messages

If you unlock the result of a free tool, we collect your email address, the page that sent you to us, and, only if you tick the optional box, your consent to marketing email. If you write to support, we keep your messages and the contact details you give us. If you ask our assistant to pass a question to the team, the email address and message you type go to our support mailbox and are kept like any other support message. What you type to the assistant before that is not stored by us.

2.5 Connected AI Apps, API Keys, Webhooks and Live Links

If you connect an AI app such as Claude or ChatGPT to PerfBee, or create an API key, we store the app's name, who connected it and when, and a hash of each key and sign-in token (never the key or token itself). For each tool call we log the time, the tool used and whether it worked. We do not receive your conversation with the app.

If your own code calls our REST API with a key, we store when the key was last used (at most once a minute). Our server logs hold, for each call, the time, the address it came from, the method, the path with its query string, the status and the user agent. We do not log request bodies. To enforce rate limits we count calls per key and per workspace, and keys that do not work per network address, each for one minute.

If you add a webhook endpoint, we store its address, the events you chose, your note and its signing secret, which is kept encrypted. For each delivery we keep the event, the time, the result (the HTTP status or a short error) and how long it took, for 14 days. The body of a failed delivery is kept for 3 days so that you can send it again. The body of a delivered event is not kept. A body holds what the event reports, such as the site address, the crawl id and counts, and it goes to the address you gave us.

If you make a live link, we store its name, the site and table it serves, its filters, who made it, and when and how often it was read. We keep a keyed hash of the link, never the link itself. Anyone who has the link can read that table. In a Google Sheet, Google fetches the table for the sheet.

2.6 Website Analytics (With Your Consent)

If you allow statistics cookies, Google Analytics records the pages you view, the buttons you click, your browser and device, and an approximate location derived from your IP address, linked to a cookie ID.

2.7 RUM Script Data (Your Visitors)

The PerfBee Real User Monitoring script you embed on your website collects only web performance data:

  • Core Web Vitals and loading times (LCP, CLS, INP, FCP, TTFB)
  • Browser name, device category (mobile or desktop) and connection type
  • The visitor's country, which our network provider derives from the IP address; the IP address itself is not stored
  • The address of the page, exactly as your website sends it, including any query string. Do not put personal data in page URLs.
  • The largest element of the page, URLs of resources that failed to load, and JavaScript error messages

The RUM script does not set cookies and does not collect user IDs, names or email addresses.

2.8 Verified Sites and Staging Access

If you verify a site, we store which site, how (a DNS record, a file or a meta tag) and when, and we check it again every week. If you save a user name and password or request headers for a verified site, we store them encrypted and use them only for requests our crawler sends to that site. They are never shown again, and never included in share links, exports or answers to AI apps. Deleting them, or your account, removes them.

2.9 Google Search Console and Analytics Data

Where it comes from. From Google, and only after you connect a Google account in Settings. Signing in with Google does not connect it. Only the workspace owner can connect or remove an account.

What we read. We read it with read-only access. First, the list of your Search Console sites, so you can choose the one that matches a site in your PerfBee account. Then, for that site, the address, clicks, impressions and average position of each page, the daily clicks and impressions of the whole site, and the index status Google reports for a page we check. If you also connect Google Analytics, we read the list of your Analytics properties, their web streams and the host names they measure, so you can choose the one that measures your site. Then, for that site, the sessions, key events and revenue of each landing page, the daily totals of the whole site, and which sessions came from AI assistants. Search queries are read only after a separate agreement, described next.

Search queries and AI features: a separate agreement. Connecting a Google account does not let us read search queries. It does not let us send your Google data to an AI model either. Both need a second agreement. Only the workspace owner can give it. The agreement belongs to one connected Google account. If you connect another account, the owner agrees again for that account. The owner can give it in Settings, Integrations, on the Google card of a site in My Sites, in Page Detail next to the search numbers of a page, or the first time an AI feature asks for it. The owner can withdraw it in Settings, Integrations or on that card at any time. It covers Search Console data only. Until the owner agrees, we read no search query and send no Google data to any AI model.

What we read with that agreement. Only with that agreement, we read the search queries people typed into Google when your pages appeared in its results, with the clicks, impressions and average position of each query on each page for the last 28 days. We also read its clicks and impressions for the 28 days before. We drop a query that contains an email address or a long number, a query that is a long pasted text, and a query that very few people searched. For a large site Google returns only part of the rows, so we may hold only the queries with the most clicks.

What we do with it, with that agreement. First, we show the queries to you in PerfBee, next to the page they belong to. Second, we use them with your page numbers in AI features that help you improve your own pages, for example a suggestion for the title of a page or a short summary of what to fix first. Some AI features run on their own, for example a short list of what to fix first that we write after each full crawl of a site. They use the summary only while the owner's agreement stands. For these, and only with that agreement, we send an AI model a compact summary of the Search Console data of the pages the feature works on. It holds page addresses, clicks, impressions and average position, the clicks and impressions for the 28 days before, and the queries of each page the feature works on, with the same figures. The summary can also include counts and a few example pages and queries from the lists in the Content tab. The feature sends the page's own title, description, headings and text along with it and, for the SEO brief, the site's open issues. We send no email address, no visitor identifier, no access token, no index status and no Analytics data. We use the queries for nothing else.

Which AI model. Our own paid account with the Google Gemini API. At present these features run only on that account. If we add support for your own AI provider key (see section 2.11), we will use it for this only after you have told us that the key belongs to a paid plan. The terms of Google's paid Gemini service say that Google does not use what we send to improve its products. Google may keep it for a limited time only to detect abuse and meet legal duties. An unpaid key is different: the provider may use what is sent through it, and a person may read it. So we never send Google data through a key that is not confirmed as paid. We never use Google data to develop, improve or train any AI or machine-learning model, ours or anyone else's.

What AI features keep. An AI feature keeps its answers, such as a suggested title, with your site, like other results. They are made from Google data, so the same rules cover them: we delete them when the agreement is withdrawn.

Withdrawing the agreement. The owner can withdraw it at any time. We then stop reading search queries, delete every search query we stored for the sites linked through that Google account, delete the answers AI features made from them, and send no Google data to an AI model. The page numbers stay, because they do not depend on the agreement. If we change these words in substance, an earlier agreement stops counting until the owner agrees again.

What we never read. We never read search queries unless the workspace owner has agreed, as described above. We never read visitor identifiers, or user, demographic or device data. We never read the query string of a page address. We never change anything in your Google account.

What we store. The data listed above, which property you linked to which site, the email address and ID of the Google account you connected, the permissions Google says you gave us, and the refresh token that lets us ask Google for the data. After the owner agrees we also store the search queries, the date of the agreement and who gave it. The token is kept encrypted (see section 11). If the workspace owner connects with a service account key from their own Google Cloud project, we store that account's email address and its key in place of the refresh token, encrypted the same way and never shown again. How long we keep the rest is in section 5.

What we use it for. Only to show it to you in PerfBee: next to each page in Site Pages and Page Detail, in the checks that flag pages Google has not indexed, in the order of your fixes in Fix Planner, in Search Opportunities, Content Intelligence and Impact Reports, and in the files you download yourself. Analytics numbers also appear in the AI Readiness and Conversion sections of your scan report. We use it for nothing else, except the AI features described above, which need the owner's agreement.

Who can see it. Only the people in your workspace. We never sell it. We never use it for advertising, and we never give it to advertising platforms, data brokers or resellers. We never send it to an AI model unless the workspace owner has agreed, as described above. We never send Google Analytics data to an AI model. Our AI insights and the custom AI prompts you run in Site Pages never receive Google data. It is never returned by the tools of our MCP server, so the AI apps you connect never receive it. Our REST API and our webhooks never carry it, and it is never put in a notification. It never appears on shared report pages. Apart from the AI step above, we do not transfer it to anyone else, except where the law requires it or where it is needed for security.

People at PerfBee. They do not read it. The exceptions are security, compliance with the law, and aggregated figures that we use for our own internal operations.

Limited Use. Our use of information we receive from Google APIs follows the Google API Services User Data Policy (opens in a new tab), including its Limited Use requirements. This covers the data we receive and the data we derive from it. Sending a summary of Search Console data to an AI model, as described above, is part of providing the AI features of PerfBee, including those that run on their own, and we do it only after the workspace owner has agreed. We do it only while that agreement stands.

Remove access. You can disconnect a Google account in PerfBee at any time, in Settings, Integrations. We then revoke our access at Google and delete the data we hold from it. For a service account we delete our copy of its key instead: the key itself stays valid until you delete it in your Google Cloud project. We delete any search queries and the answers AI features made from them as well. You can also remove PerfBee yourself on the Google account permissions page (opens in a new tab). After that we cannot read anything new. We hide the numbers we hold at once, and we delete them 14 days after Google reports that our access is gone, unless you connect the account again before then. You can also ask us to delete them sooner (see section 8).

2.10 Browser Extension

If you install the PerfBee Content Monitor Picker for Chrome, it does nothing on a page until you click its icon and choose to pick an element. It does not read pages in the background, and it does not collect your browsing history.

When you sign in from its popup, your email address and password go to PerfBee's sign-in, as on the website. The extension does not keep the password. It keeps your email address and a session token in its own storage in your browser. Signing out removes both. If the session expires, the extension removes the token and keeps the email address, so you can sign in again.

When you open its panel on a page, the extension sends us the domain of that page, so we can show the rules and page types you already have for it. When you save a rule, it sends us the domain of the page, the name you gave the rule, the selectors that find the element, the settings you chose (such as the page type and the expected count) and, if you chose a text match, the piece of text you matched on. To check that our crawler can see the element, it also sends the address of the page, including its query string. We fetch that page once as our crawler. The extension sends nothing else from the page.

We keep a rule like the rest of your account data (see section 5). We use all of it only to run your Content Monitor rules. We do not sell it, and we do not use it for advertising.

2.11 Your Own AI Provider Key

If you add your own key for an AI provider (OpenAI, Anthropic or Google Gemini) and run an AI prompt in Site Pages, we send your instruction and the page fields you ticked for each selected page (the address, title, meta description, H1, word count, canonical URL, meta robots, HTTP status, declared language, first H2, social title and description, a summary of its structured data and the page's main text, as Site Pages stored them) to the provider you chose, using your key. If you ask a question about one page in Page Detail, we send your question and all of those fields of that page; if you ask us to read the live page, or no page text is stored, we fetch the page once as our crawler and send what we read from it in place of the stored values, without storing it in our database. We send nothing else, and in an AI prompt we never send data from Google Search Console or Google Analytics. Apart from the case in the next paragraph, we send nothing to a provider at any other time. The provider handles that text under your own agreement with it and charges your account. We do not choose or control it, so it is not in the service providers table in section 6. We keep the key encrypted and show only its last four characters to the workspace owner and admins. Removing the key deletes it. We keep each answer of an AI prompt (up to 6,000 characters, with the code in it once we have checked it) with its page until you delete the prompt, the site or your account, or until the page leaves Site Pages. A question asked in Page Detail is deleted once it is answered; the answer is kept for 10 minutes so that your screen can show it, then deleted, and neither is stored in our database. We do not keep the provider's full reply.

Your key and Search Console data. At present no AI feature sends Search Console data through your key. They run only on our own paid Gemini account (see section 2.9). If we add support for your key, we will use it for the summary described in 2.9 only if the workspace owner has agreed (see section 2.9). The owner or an admin must also have told us in the key's settings that the key belongs to a paid plan. We would accept this only for a provider whose terms we have checked, at present Google Gemini. We keep that statement with the key and delete it with the key. We cannot check it with the provider, so it must be true. An unpaid key can let the provider use what is sent through it and let a person read it, so we never send Google data through a key that is not confirmed as paid.

2.12 Server Log Files (Bot Tracker)

If you upload a server log to Bot Tracker, we read it to count the visits of search engine and AI crawlers to your site. A log line can hold a visitor's IP address, the page they asked for and their browser. The file waits on our server until it is read. We process each line in memory, and we delete the file when processing ends, whether it worked or not. If something goes wrong, a leftover file is deleted within two hours. The screen shows a command that keeps only the crawler lines, so that the lines of your human visitors never leave your server.

What we keep. Counts. For each day, crawler and page, we keep how many visits there were and how many got each kind of answer (success, redirect, refused with a 401, 403 or 429, not found, another 4xx, 5xx). For a page we keep its path and, for an address with a query string, only the names of the parameters, never their values. A path is stored as it appears in your log, so it can contain what your site puts in its URLs. For each upload we keep a record: when it arrived, who uploaded it, its format and size, how many lines were read and kept, the days it covers, a hash of the file, and a result.

What we do not keep. IP addresses, browser (user agent) strings, referrers and log lines. Lines that do not come from a crawler we read are counted as a number and then discarded. A visit that claims to be a crawler but fails our check is counted without a page address.

How we check a crawler. While we read the file, we compare the IP address of each crawler visit with the list of addresses that the crawler's operator publishes, in memory. The address is not stored. The lists are fetched from the operators' own websites with a plain request that carries nothing from you or your log. No new service receives personal data, so the table in section 6 has no new row.

For this data you are the controller and we are the processor. Upload only logs that you may use for this purpose (see our Terms of Service). How long we keep the counts is in section 5.

2.13 Cloudflare Analytics (Bot Tracker)

Where it comes from. From Cloudflare, and only after the workspace owner connects a site's Cloudflare zone in Bot Tracker with an API token they create in their own Cloudflare account. Only the workspace owner can connect, test or disconnect it.

What we read. We only ask Cloudflare questions. We look up the zone that holds the site's address. Each question carries the token, the host name of the site and the names of the ten crawlers. Then, once a day, we read how many requests came from search and AI crawlers on each of the days we have not read yet, grouped by page address (without the query string), crawler user agent and response status group. We read back as far as your plan keeps its history and Cloudflare still has the data.

What we never read. We never read IP addresses, visitor data, or the content of any request or page, and we never change anything in your Cloudflare account.

What we store. The token, encrypted under a key of its own and never shown again; the name of the zone, the state of the connection, when we last read and, if your plan no longer includes Bot Tracker, the day we noticed; and, for each day, the number of requests of each crawler to each page address, by status group. We keep the name of the crawler, not the user agent text, and no IP address and no single request. A user agent is something any visitor can send, so these visits are shown as "user agent only" and never as verified. For the same reason, we also keep the page address of a visit that only claims a crawler name. Cloudflare may sample busy days, and then the counts are estimates.

What we use it for. Only to show you the crawler visits of your site in PerfBee. We never sell it and never use it for advertising. It is visible to the people in your workspace.

Remove access. Disconnecting in Bot Tracker deletes the token and every count we read from Cloudflare at once; the page addresses that only those counts used follow at the latest in our nightly clean-up. Removing or switching off the site, changing its address to one the zone does not hold, or closing your account does the same, at the latest in our nightly clean-up. If you delete the token in Cloudflare, we can read nothing new, and the counts we already read stay until they pass your plan's history limit, or until you disconnect. If your plan changes so that it no longer includes Bot Tracker (a plan change, an ended subscription or an ended trial), we stop reading at once and keep the token and the counts for 30 days. If the plan has not come back by then, we delete the token and every count we read from Cloudflare, exactly as if you had disconnected. If it comes back before then, nothing is deleted.

3. How We Use Your Data

  • Service delivery: Running audits, crawls, and RUM analytics on your behalf.
  • Google data: Showing you the numbers you connected from Google next to your pages, and ranking your issues by them. If the workspace owner agrees, also reading search queries and using Search Console data in AI features. Nothing else (see 2.9).
  • Billing: Processing payments and managing your subscription.
  • Communication: Sending service emails (alerts, scan results, account and billing notices) and setup tips for new accounts. Payment receipts come from our payment provider, Paddle.
  • Support: Responding to your requests and troubleshooting issues.
  • Bot Tracker: Counting crawler visits in a server log you upload or in the Cloudflare analytics you connect, and showing you the result (see 2.12 and 2.13). Nothing else.
  • Security: Detecting fraud, abuse, and unauthorized access.
  • Website analytics: Understanding how our website is used, only with your consent.
  • Product improvement: Analyzing aggregate, anonymized usage patterns to improve features.

We do not use your data for advertising or profiling, and we do not sell it to third parties.

For users in the European Economic Area (EEA) and United Kingdom, our legal bases are:

  • Contract (Art. 6(1)(b)): Processing necessary to deliver the service you subscribed to.
  • Legitimate interest (Art. 6(1)(f)): Security monitoring, fraud prevention, service improvement, our support assistant, and setup emails that help new users start using the service they signed up for.
  • Consent (Art. 6(1)(a)): Product news emails and statistics cookies. You can withdraw consent at any time.
  • Legal obligation (Art. 6(1)(c)): Tax and financial record-keeping.

5. Data Retention

  • Account data: retained for the duration of your subscription plus 90 days after closure. The Google connection is the exception: when you close your account we revoke PerfBee's access at Google immediately and delete the Google data with it.
  • Data from Google Search Console and Google Analytics: the page numbers are replaced on every refresh, so only the latest page numbers are kept. For a page you rewrite with AI Page Rewriter, we also keep its Search Console numbers and its Analytics sessions and key events for the days before and after the change, so Impact Reports can compare them. If you unlink only Google Analytics, we delete the Analytics part. Daily totals are kept for 16 months. URL inspection results are kept for 90 days. All of it is deleted when you disconnect the Google account, unlink a site or close your account. It is also deleted 14 days after Google reports that our access is gone (see 2.9), unless you connect the account again before then.
  • Search queries, and the answers AI features made from Google data (only after the workspace owner agrees, see 2.9): the queries are replaced on every refresh in which Google returns queries, so only the latest are kept. If Google returns none, or cannot answer, we keep the earlier queries and replace them at the first refresh that returns some. We delete both at once when the owner withdraws the agreement, and with the rest of the Google data in every case listed above.
  • Performance and RUM data: retained according to your plan's history limit (30 to 365 days).
  • Large export files (built in the background when an export is too large to build while you wait): kept for up to 72 hours so that you can download them, then deleted.
  • Billing records: retained for 7 years to comply with financial regulations.
  • Deleted data can stay in our server backups for a limited time until those backups are replaced.
  • Bot Tracker (see 2.12): the counts and the upload records are kept for your plan's history limit (at most 365 days), then deleted each night. The uploaded file itself is deleted when processing ends and in any case within two hours. Forgetting an upload removes it from your list, deletes its file and drops the link to the person who uploaded it; the counts it added, a fingerprint of the file and the time of the upload stay until they pass your plan's limit, so that the same file is not counted twice. All of it is deleted when you delete the site or your account. Large amounts are finished by our nightly clean-up, at the latest the next night.
  • AI provider keys you add: kept encrypted until you remove them or close your account. AI prompt answers: kept with their page until you delete the prompt, the site or your account, or until the page leaves Site Pages (its scans pass your plan's retention, or it has been missing from your scans for longer than that).
  • Cloudflare connection (Bot Tracker): the token is kept encrypted until you disconnect, change the site's address to one the zone does not hold, remove or switch off the site, or close your account, or until 30 days after your plan stops including Bot Tracker unless it comes back before then; if we cannot delete it at once, our nightly clean-up does, at the latest the next night. The daily counts read from Cloudflare are kept for your plan's history limit (at most 365 days) and then deleted, and all of them are deleted when you disconnect or when that 30 days pass.
  • Support messages (including those sent through the assistant's email form), and records of your cookie choices: kept as long as needed to handle your request or to prove your consent.
  • Site access details (staging user names, passwords and headers): kept until you delete them or your account.
  • Connected apps and API keys: listed in your account until you delete it. A disconnected app or revoked key stays in the list, marked as such, and can no longer be used.
  • Webhook endpoints: kept until you delete them or your account. Delivery records: 14 days. The body of a failed delivery: 3 days.
  • Live links: listed in your account until you delete it. A revoked or expired link stays in the list, marked as such, and can no longer be read.
  • Anonymized aggregate data: may be retained indefinitely for statistical purposes.

6. Data Sharing & Sub-processors

We share data only with the service providers below, under data processing agreements or the provider's standard terms. We do not share data with advertising networks or data brokers.

ServiceProviderPurposeDataDetails
HostingContabo GmbHHosts the platform and its databaseAll data you give us and the data we create for you
CloudflareCloudflare, Inc.Content delivery, DDoS and bot protection, including the human check in our chat assistant (Turnstile)IP address, browser and request data
PaddlePaddle.com Market LimitedPayments, invoices and tax, as Merchant of RecordName, email, billing address and payment details (we never see your card number)
PostmarkAC PM LLC (ActiveCampaign)Sends our email: system emails, setup tips and, for people who opted in, product newsEmail address and the content of the email
Gmail (Google Workspace)GoogleOur mailboxes: support email and replies to our emailsEmail address and the content of the email
Google Gemini APIGoogleAI insights and code-fix suggestions; finding the prepared answer to a question typed into the support assistant on our website; and, only after the workspace owner agrees (see section 2.9), AI features that use a summary of your Search Console dataURLs, page and performance data of the sites you analyze; the text of a message you type into the support assistant (we do not store it); with that agreement, page addresses, clicks, impressions, average position and search queries from Search Console, together with the title, description, headings and text of the page and, for the SEO brief, the open issues of the site; never Analytics data
Google PageSpeed InsightsGoogleSpeed tests of pages you submitThe URL you test
Google Search Console and Analytics APIsGoogleReads the search and traffic numbers of the sites you connect, and the index status of the pages you inspect, only after you connect your own Google account. This is not our website statistics (see Google Analytics and Tag Manager below)Page URLs, clicks, impressions, average position, sessions, key events, revenue and index status. Search queries only after the workspace owner agrees (see section 2.9). No visitor identifiers. Google receives the page URLs you inspect. We receive the email address of the Google account you connect
Google Analytics and Tag ManagerGoogle Ireland LimitedWebsite analytics and tag management; analytics only after you allow statistics cookiesPages viewed, button clicks, browser and device, approximate location from your IP address, cookie ID
Google FontsGoogle Ireland LimitedDelivers our icon fontIP address and browser information
CookiebotUsercentrics A/SRecords and applies your cookie choicesYour consent choices and the technical data needed to prove them
Cloudflare analytics (your own account)Cloudflare, Inc.Reads how often search and AI crawlers requested the pages of a site, only after the workspace owner connects the Cloudflare zone of that site with a token they createThe token, the host name of the site and the names of the ten crawlers go to Cloudflare with each request. We receive daily counts by page address, crawler user agent and response status group. No IP addresses, no visitor data, no query strings
jsDelivrjsDelivrDelivers an open-source charting library to the dashboardIP address and browser information
TrustpilotTrustpilot A/SThe review box on the dashboard home, shown to account owners and team members until they hide it, so you can rate PerfBee on TrustpilotYour name, email address and PerfBee account number, encrypted, sent when the review box loads; the address of the dashboard page, IP address and browser information. A review you write there falls under the privacy policy of Trustpilot
Sign-in providersGoogle, GitHub, ShopifySign-in, only if you choose itThe name and email address they share with us

When you connect an AI app such as Claude or ChatGPT, the results of the tools you call go to that app and its provider under your own agreement with them. We do not choose or control that provider, so it is not listed above. The same is true of the address of a webhook endpoint you add, and of a Google Sheet or any other tool that reads a live link you made.

We may update our sub-processors from time to time and will notify customers of any material changes. A full sub-processor list is available upon request at support@perfbee.com.

7. International Data Transfers

Our primary servers are hosted in the European Union. Because we are based in Türkiye, our team also accesses data from Türkiye, and several providers listed in section 6 (for example Google, Cloudflare and Postmark) may process data in the United States or other countries outside the EEA. Where this happens, we rely on an adequacy decision (such as the EU–US Data Privacy Framework for certified providers), on the Standard Contractual Clauses approved by the European Commission, or on equivalent safeguards. You may request more information at support@perfbee.com.

8. Your Rights

8.1 GDPR Rights (EEA / UK Residents)

  • Right of access (Art. 15)
  • Right to rectification (Art. 16)
  • Right to erasure / “right to be forgotten” (Art. 17)
  • Right to restriction of processing (Art. 18)
  • Right to data portability (Art. 20)
  • Right to object (Art. 21), including to direct marketing
  • Right to withdraw consent at any time (Art. 7(3))
  • Rights related to automated decision-making (Art. 22) — see section 12

8.2 CCPA Rights (California Residents)

  • Right to know what personal information is collected
  • Right to delete personal information
  • Right to opt-out of sale (we do not sell personal information)
  • Right to non-discrimination for exercising CCPA rights

To exercise any of these rights, email support@perfbee.com. We will respond within 30 days (GDPR) or 45 days (CCPA). We may verify your identity via your registered email address before fulfilling requests.

9. Cookies

We use necessary cookies to keep you signed in and to remember your cookie choice. With your consent we also use Google Analytics (statistics). Our support assistant sets no cookies of its own. Its human check, run by Cloudflare (Turnstile), may keep one small item in the local storage of your browser. We do not use advertising cookies. The full list, and a button to change your choice at any time, are in our Cookie Policy.

10. Direct Marketing

We send two kinds of email besides system email.

Product news: about once a month, only if you ticked the optional box at sign-up, in a free-tool form, or later in Settings. We do not send product news to recipients in Türkiye until our registration with the national İYS system is complete.

Setup tips: after you register we may send up to three emails about your own account and the next step you have not done yet, for example adding your first site or scheduling a scan. They contain no offers and stop once you finish the step. We rely on our legitimate interest in helping new users use the service they signed up for and, where it applies, on the ePrivacy rule for messages to existing customers.

Every product news and setup tip email has an unsubscribe link, and you can turn both off in Settings. This never affects system email. System emails (account, security, billing, trial and payment notices, and the scan results and usage alerts you have switched on) are not marketing.

11. Security & Data Breaches

We protect your data with encryption in transit (TLS), hashed passwords, encrypted storage of the tokens of connected Google accounts under a key of their own, role-based access control, and a firewall with intrusion prevention on our servers. Access to personal data is limited to the people who need it to run the service.

If a data breach puts your rights and freedoms at risk, we will notify the competent supervisory authority within 72 hours of becoming aware of it, and affected individuals within 3 days, by email and with a notice in the dashboard. The notice will say what happened, which data was affected and what you can do.

The token of a Cloudflare connection in Bot Tracker is kept encrypted as well, under a key of its own.

12. Automated Decision-Making

PerfBee does not perform automated decision-making or profiling that produces legal or similarly significant effects on individuals, as described under Art. 22 GDPR. Platform features (e.g., anomaly alerts, credit limits) are rule-based and do not constitute automated individual decision-making in the legal sense.

13. Data Processing Agreement (DPA)

As a data processor for RUM visitor data collected on your behalf, we are prepared to enter into a Data Processing Agreement (DPA) with customers subject to GDPR. Customers may request a DPA by contacting support@perfbee.com.

14. Children’s Privacy

PerfBee is a B2B service intended for businesses and developers aged 18 and over. We do not knowingly collect personal data from anyone under 18. If we discover that such data has been collected, we will delete it promptly.

15. Changes to This Policy

We may update this Privacy Policy from time to time. We will notify you by email and by posting a notice in the dashboard at least 30 days before material changes take effect. A change that only adds a feature and the data it uses takes effect on the day we publish it. That day is the “Last updated” date at the top of this policy. Continued use after the effective date constitutes acceptance.

16. Contact & Complaints

For privacy questions, DPA requests, or to exercise your rights: support@perfbee.com

If you believe we have not handled your data appropriately, you have the right to lodge a complaint with your local supervisory authority. In the EU, find your authority at edpb.europa.eu. In the UK, contact the ICO at ico.org.uk.