Privacy Notice Version 1.10 Controller: David Klimek Zanetic (Vidljivo). We use your sign-up details to create and manage your account and keep it secure. To do the work you ask for, your business content is processed by AI model providers, some outside the European Economic Area, on standard contractual clauses. We ask them not to retain it or train on it, except when generating videos; Google may keep pictures, videos and their requests to detect abuse. We do not sell your data. The Privacy Notice lists the recipients, the transfers, the retention periods and your rights (access, rectification, erasure, restriction, portability, objection, and complaint to a data-protection authority). Accepting the Terms is not consent to data processing. Optional Google Analytics on the public website requires a separate visitor choice; rejecting it does not restrict use of the service. It is independent of connected-business and AI-sharing permissions. Connected-service report context is kept for up to 30 days; supported dated historical figures for up to 13 months. 1. Who is responsible The controller of your personal data is: Provider: David Klimek Zanetic Trade name: Vidljivo Tax ID (NIF): 26591041M Address: Carrer Vinya del Rei 33, Alella, 08328 E-mail: hello@vidljivo.com Status: self-employed individual (autónomo), Spain Complaints: hello@vidljivo.com, or by post to the address above. Not adhered to any alternative dispute resolution (ADR) body. You can write to hello@vidljivo.com for any privacy question or request. Messages to that address are answered by Vidljivo’s AI agents, under the responsibility of the provider, as the Terms of Service (vidljivo.com/legal/terms) explain. Requests about your personal data, formal complaints and legal matters are handled by the provider personally, never by an AI agent: the agent that receives such a message only acknowledges it and hands it on. 2. Our roles For the data of your own website, your customers and your reviews that you give us to work on, we are your processor and the Data Processing Agreement (vidljivo.com/legal/dpa) applies; you are the controller. For your account, security, billing and invoicing, support, and the pseudonymised events that let the agents that oversee the service learn, we are the controller, and this notice applies. 3. Data we process • Account data: name, e-mail address, sign-in and security records (with your Google or Facebook account number if you sign in with one), language, the version of the Terms you accepted (with the size and fingerprint of the exact text) and when, and your separate optional choices. • Business data you provide or connect: websites, brand details, content, pictures and videos, and information from the Google, social, advertising or website accounts you choose to connect, and the record of your ads’ results (section 6). • Work records: scans, findings, proposals, approvals, jobs, change history and your conversations with the agents in the team chat of the app, with the notes the agents keep from it when chat memory is on (section 5). • Payment and contract data: payments are handled by Stripe and invoices by Quaderno; we receive the plan, the status, the invoices, the tax details and the last digits of the card, not the full card number. • E-mails you send us, with their attachments, our replies, and the support messages and feedback you give. • Technical and security logs, identifier-free service counters and, only with your separate choice, pseudonymous public-website Analytics data (section 9). • A salted hash of the network prefix of your connection (the IPv4 address, or the IPv6 /64 and its /48 group), never the address itself, kept per account and copied onto the usage records of the AI team, to limit abuse of the allowance that all Free accounts share (sections 4 and 7). • A conversation you choose to send to customer support, with your account e-mail address (section 11). 4. Why we use it and on what basis • Purpose: Create and run your account and provide the service you order; Data: Account data, business data, work records; Legal basis: Performance of the contract (art. 6(1)(b) GDPR) • Purpose: Take payment, issue invoices and keep the records the law requires; Data: Payment and contract data; Legal basis: Performance of the contract; legal obligation (art. 6(1)(c)) • Purpose: Keep the service secure and prevent abuse; Data: Security logs, account data; Legal basis: Legitimate interest (art. 6(1)(f)) and legal obligation • Purpose: Limit abuse of the allowance that all Free accounts share; Data: A salted hash of the network prefix, per account; Legal basis: Legitimate interest (art. 6(1)(f)); no recipient • Purpose: Keep evidence of the documents you accepted and the choices you made, and defend legal claims; Data: Acceptance records; Legal basis: Legitimate interest; legal obligation • Purpose: Answer your requests and support messages, including a conversation you send to support; Data: Support data, e-mails; Legal basis: Performance of the contract; legitimate interest; your request • Purpose: Send you e-mails that help you use the service: approvals waiting, set-up and how-to steps, a weekly summary; Data: Account data, the state of your brand’s work; Legal basis: Legitimate interest; you can stop them in Settings or with the link in each e-mail • Purpose: Send you offers, for example a plan that fits your brand better; Data: E-mail address, plan; Legal basis: Your consent, given separately; you can withdraw it at any time in Settings or with the link in each offer • Purpose: Keep notes from the team chat, when chat memory is on; Data: Your own words in the team chat; Legal basis: Performance of the contract; you can remove each note • Purpose: Keep the record of your ads’ results, and give similar brands a combined summary; Data: The record of your ads’ results (section 6); Legal basis: Performance of the contract for your own record; legitimate interest for the combined summary; both only while you share platform numbers • Purpose: Optional Google Analytics on the public marketing website; Data: Pseudonymous browser/session and public-page data as described in section 9; Legal basis: Your separate visitor consent; rejecting it does not restrict the service • Purpose: Understand where people stop and improve the service; Data: Daily counters without an identifier; Legal basis: Legitimate interest • Purpose: Let the agents that oversee the service learn; Data: Pseudonymised events and outcomes; Legal basis: Legitimate interest • Purpose: Use raw text beyond your own account (tier 3); Data: Raw text after redaction and human review; Legal basis: Your consent, which you can withdraw at any time; switched off • Purpose: Process the data of your website, customers and reviews for you; Data: Customer personal data; Legal basis: Your instructions, as your processor (the legal basis is yours) Accepting the Terms is not consent to data processing; the processing above relies on the bases stated. We do not sell personal data. No decision with legal or similarly significant effects on you is taken solely by automated means: AI proposals are carried out only within the approvals you set. 5. Artificial intelligence and recipients The agents use AI models that we reach through OpenRouter, Inc., a router that sends each request to a model host. We use a closed list of model providers; if none of them can serve a request, the request fails and nothing is sent to any other provider. With every chat request, including a request to describe an image, we ask the provider not to retain it and not to use it for training. That is a request, not a guarantee: a provider’s handling of data is governed by its own terms with OpenRouter and with us. A request to generate a picture carries the same request. A request to generate a video cannot, because no video model we can reach offers it, and OpenRouter keeps a generated video for a short time so that we can download it. Google, which generates the pictures and videos, may keep a request and what it generates to detect, investigate and prevent abuse and, under the terms that apply to it, does not use them for training; we do not promise how long it keeps them. Such a request is a description of the picture or video that our agents write from your brand’s public details; it is not built from the reports of the platforms you connect or from your customers’ data, but it can include your brand name, which is personal data when it is a person’s name. OpenRouter is a United States company, so requests pass through the United States, and we do not claim that processing stays in any region. OpenRouter may use a small sample of request inputs for anonymous categorisation by a model it hosts, as its terms state. When the agents may use a photo from your Library, Kiara can send it to Google’s image model to make an improved copy. Your original stays unchanged, the copy is saved in your Library, and an Instagram post that uses it is labelled as made with AI. We don’t send a photo if Kiara sees a person in it. Incoming e-mail is processed by the same AI providers under the same settings. Chat memory. When chat memory is on, the agent who answers you in the team chat may keep a short note of something lasting you said: a goal, a limit, a preference, or a fact about your customers or your business. You see the note in the answer and in Settings, where you can remove it. The agents treat a note as what you said, not as a proven fact, and never as permission to publish, spend or approve. We do not keep a note that holds contact details, long numbers, passwords, or sensitive facts about a person, such as health or religion. Notes are sent to the AI providers with the other context of a task. • Recipient: OpenRouter, Inc. Role: AI router and processor; a US company, so requests pass through the United States; Country: United States (New York); Transfer safeguard: Standard contractual clauses (Module 2, Irish law) with the UK addendum, in OpenRouter’s data processing addendum of 26 August 2026; What it receives: Each request sent to an AI model: your question and the context the task needs • Recipient: Google Analytics (optional public-website measurement); Role: Separate analytics recipient, only after visitor consent; Country: May be processed outside the EEA, including the United States; Transfer safeguard: Applicable Google terms and safeguards; no new transfer agreement is inferred from other processor contracts (section 9); What it receives: Public-page events, pseudonymous browser/session identifiers and network/device data; not private app content • Recipient: Google (Google Cloud, Vertex AI); Role: Model host for the Anthropic models we use, and provider of the Gemini image-generation and Veo video-generation models we use; processor; Country: United States; Transfer safeguard: Standard contractual clauses; What it receives: The requests routed to its models • Recipient: Microsoft (Azure); Role: Model host for the OpenAI models we use; processor; Country: United States; Transfer safeguard: Standard contractual clauses; What it receives: The requests routed to its models • Recipient: Amazon Web Services (AWS Lightsail); Role: Hosting and encrypted backups; processor; Country: Germany (Frankfurt); Transfer safeguard: None needed: processing inside the European Economic Area; What it receives: Everything the service stores, in encrypted form for backups • Recipient: ZeptoMail (Zoho, EU data centre); Role: Sends the service’s e-mail and the agents’ replies; processor; Country: European Union; Transfer safeguard: None needed for the EU data centre; group transfers on standard contractual clauses; What it receives: The e-mail we send you: address, subject and text • Recipient: Zoho Mail (EU data centre); Role: Receives and stores the e-mail addressed to Vidljivo; processor; Country: European Union; Transfer safeguard: None needed for the EU data centre; group transfers on standard contractual clauses; What it receives: The e-mail you send us, with its attachments • Recipient: Stripe; Role: Payment service provider; Country: Ireland (Stripe Payments Europe), with group companies in the United States; Transfer safeguard: Standard contractual clauses for transfers within the group; What it receives: Name, billing address, card data you enter at Stripe, amount and plan • Recipient: Quaderno; Role: Issues invoices and calculates the tax in them; processor; Country: European Union; Transfer safeguard: None needed for the EU; What it receives: Invoice data: name, billing address, tax ID, plan, amount and tax Transfers to the processors described above outside the European Economic Area rely on the European Commission’s standard contractual clauses. Optional public-website Google Analytics is a separate purpose and recipient described in section 9; we do not infer an Analytics transfer agreement from those processor arrangements. Payment data goes to Stripe to take payments; Stripe and Quaderno act for the purpose of payment and invoicing, and the platforms you connect receive only what you ask us to send them. Authorities receive data only where the law requires. We do not sell data or share it for advertising by third parties. The agents that oversee and improve the service learn in tiers. Tier 1: your own agents read your own raw text and history to do your work, and other clients’ agents never see it. Tier 2: the overseers learn only from pseudonymised events and outcomes (for example, that a kind of action succeeded), not from your raw text; pseudonymised events are still personal data and are covered by your rights. Tier 3: raw text goes to the overseers only with your separate, optional opt-in, after redaction and human review; this tier is switched off. Credentials of connected accounts are never exported in any tier, and neither are conversations or prompts, except raw text under tier 3. 6. Google, Meta and TikTok We describe here what we use each platform’s data for. We ask each platform for the permissions of one step at a time and read first. Declining a permission turns off only the feature that needs it. Platform data is not sold, is not used to train models, and is shared only with the recipients of section 5 that the task needs. No person at Vidljivo reads it except with your consent, to keep the service secure, to comply with the law, or as aggregated figures. Summaries of the figures we read from Google, Meta and TikTok reach the AI providers of section 5 only if you switch on the permission to share platform numbers with the agents. That permission is off by default; with it off, no figure from these platforms is sent to a model. The record of your ads’ results. While you share platform numbers, we keep a record of each ad that Kora prepared and you chose, on Meta or Google. It holds the plan, the daily cost, views and clicks, and, where Kora gave one, whether the ad reached the range she expected. Kora uses it to plan better ads for you. It holds no enquiries or sales. Daily figures are kept for 30 days and then added to the totals. An ad’s record is kept up to 13 months after its last measured day. When you stop sharing platform numbers, the whole record is erased at once. When you disconnect Meta Ads or Google Ads, or remove Vidljivo’s access there, it is erased 30 days later, unless you connect it again before then. A connection that only expired erases nothing. Closing the brand or the account erases it. Similar brands. The finished results of your ads also help brands of the same kind of business in the same country. Their Kora sees only a summary that combines the results of brands of at least 5 other customers. It never shows your name, your website, your account or a single result. Your Kora can see such a summary in the same way. This happens only while you share platform numbers. To keep your results out, stop sharing platform numbers; that also erases your own record. Google Vidljivo’s use and transfer to any other app of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements. • Service: Sign-in with Google (permissions openid, e-mail, profile); What we read: Your name, e-mail address and Google account identifier; What we use it for: To create your account, sign you in and keep it secure; nothing else; How long we keep it: While the account is open • Service: Search Console (read-only); What we read: Search performance of the properties you pick: queries, pages, clicks, impressions, average position and indexing status; What we use it for: To show your results and to prepare proposals you approve; How long we keep it: Authenticated report context: up to 30 days; qualified dated historical cells up to 13 months where supported, without a complete-history promise. Agents keep measurement dates and source freshness distinct; current-action checks may require newer data • Service: Google Analytics (read-only); What we read: Traffic and engagement figures of the properties you pick: users, sessions, engagement rate and key events, and, where your site has them, e-commerce figures and figures per landing page (without query strings); What we use it for: To show your results and to prepare proposals you approve; How long we keep it: Authenticated report context: up to 30 days; qualified dated historical cells up to 13 months where supported, without a complete-history promise. Agents keep measurement dates and source freshness distinct; current-action checks may require newer data • Service: Google Ads (permission adwords; changes only where enabled); What we read: Campaign names, budgets, status and performance figures of the account you pick; What we use it for: To show your results and to prepare proposals you approve; where changes are enabled for your account, to make only the change you approved exactly (see below); How long we keep it: Mapped report analysis availability: 30 days; for a campaign Vidljivo created for you, dated daily impressions and clicks, and cost only where that read collected it, up to 13 months (section 7); control and approval records follow their separate periods Google Ads is read for the account you pick. Changes to that account are possible only where Vidljivo has enabled them for your account, and only the changes the app offers: a daily budget change or a pause, a new campaign created paused, the activation of that campaign, and a pause of a campaign Vidljivo created. Each is made only after you approve that exact change in the app; an activated campaign can spend its approved budget. Until changes are enabled, Vidljivo only reads the account. Disconnecting in the app ends the connection in Vidljivo and the tokens we hold are deleted from our systems within 24 hours; to make sure the access itself is gone, check the third-party access of your Google Account and remove Vidljivo if it is listed. Both are explained on the data deletion page (vidljivo.com/legal/data-deletion). Meta (Facebook and Instagram) • Permission stage: Sign-in with Facebook (public_profile, email; a separate Meta app); What we read or do: Your name, e-mail address and Facebook account number; What we use it for: To create your account, sign you in and keep it secure; nothing else; How long we keep it: While the account is open, or until Facebook asks us to remove the link • Permission stage: Reading (pages_show_list, pages_read_engagement, instagram_basic); What we read or do: The list of your Facebook Pages, the content and engagement figures of the Page you pick, and the basic details and media of its linked Instagram professional account; What we use it for: To show your accounts and results and to prepare proposals you approve; How long we keep it: Mapped report analysis availability: 30 days; no dedicated daily archive; control and approval records follow their separate periods; access tokens while connected • Permission stage: Publishing (adds pages_manage_posts, instagram_content_publish); What we read or do: Creates the posts you approve on the Page or the Instagram account; What we use it for: To publish only what you approved, item by item or under a standing authorisation (below), with the AI disclosure that applies; How long we keep it: The post stays on the platform until you remove it; our record of the approval follows the retention table • Permission stage: Replies (adds instagram_manage_comments; for a Facebook Page, pages_read_user_content and pages_manage_engagement); What we read or do: Reads the comments on your Instagram posts and your Facebook Page, with the commenter’s public name and text, and posts the replies you approve; What we use it for: To help you answer comments; no reply is sent without your approval of that reply or a standing authorisation that covers it; How long we keep it: Mapped report analysis availability: 30 days; no dedicated daily archive; control and approval records follow their separate periods • Permission stage: Insights (optional; adds read_insights for a Facebook Page, instagram_manage_insights for an Instagram account); What we read or do: Insights figures of the Page and of the Instagram account and its media; What we use it for: To show your results and to prepare proposals you approve; How long we keep it: Mapped report analysis availability: 30 days; no dedicated daily archive; control and approval records follow their separate periods • Permission stage: Meta Ads (optional; ads_read, plus ads_management in a separate enabled management stage); What we read or do: Identifiers, role/status, currency and time zone of the selected ad account; campaign and creative information, budgets, reported spend, impressions, clicks, reach and actions, and transparency payer/beneficiary details. Account-wide reports are distinguished from campaigns verified as created through Vidljivo for this brand. What we use it for: Show advertising results independently of AI sharing; use authorised summaries for agent proposals only when platform-number sharing is enabled. Supported changes require the exact approved review: paused campaign setup, separate explicit activation, pause/stop or an existing daily-budget edit. An activated ad can spend its approved budget. How long we keep it: Mapped report context: up to 30 days from its original supporting read. Supported dated figures of verified own campaigns: up to 13 months (section7), not a whole-account daily archive. Approval and action records follow their separate periods. Sign-in with Facebook uses a separate Meta app. It asks only for your public profile and e-mail address. We keep your Facebook account number to sign you in. We do not keep the access token Facebook gives us. Facebook does not tell us whether the e-mail address is confirmed, so we ask you to confirm it. If you remove Vidljivo in your Facebook settings, or ask Facebook to delete your data, we remove the link between your Facebook account and Vidljivo. That does not close your Vidljivo account: you can close it in Settings. When you ask Facebook to delete your data, we also keep a short record of that request (section 7). Connecting uses Facebook Login for Business, with separate permission stages for reading, publishing, replies and optional insights. For supported posts and eligible replies, we use your approval of the item or a current recorded authorisation for the selected brand and account, within its formats and limits. Media must pass the applicable rights checks; generated images require the covered AI-image option. You can revoke the authorisation, and other scope or eligibility changes can also stop new actions. This authorisation does not cover ads, spending or TikTok. Covered posts may publish when background preparation completes; posting at a chosen time is unavailable. Own or already-replied comments are excluded; flagged comments need individual approval. For covered AI-generated Instagram feed images, the app requests AI disclosure; Instagram controls its display. The access tokens are encrypted by the application. The other data we read from Meta is stored in our database and protected by access controls; the application does not encrypt it separately. Disconnecting in the app ends the connection in Vidljivo and the tokens we hold are deleted from our systems within 24 hours, but it does not call Meta to remove the permission you gave: check your Facebook settings and remove Vidljivo there as well. TikTok • Permission: user.info.basic; What we read: Your TikTok display name, avatar and account identifier; What we use it for: To show which TikTok account is connected; How long we keep it: While connected; the tokens are deleted from our systems within 24 hours of disconnecting • Permission: video.list; What we read: The list of your public videos: identifiers, titles, cover images and links; What we use it for: To show your recent videos next to your other channels and to prepare proposals you approve; the service does not publish to TikTok; How long we keep it: Mapped report analysis availability: 30 days; no dedicated daily archive; control and approval records follow their separate periods • Permission: TikTok Ads (separate read-only advertiser connection, where offered); What we read: Daily spend, impressions and clicks of the advertiser account you select; no campaign or conversion data; What we use it for: To show paid results next to your other channels; Vidljivo makes no change to your TikTok ads; How long we keep it: Mapped report analysis availability: 30 days; dated daily figures up to 13 months where supported (section 7) Disconnecting TikTok in Vidljivo removes the local connection only; it does not ask TikTok to end its side. To end TikTok’s access, remove Vidljivo in the TikTok app: open your profile menu, go to Settings and privacy, then Security and permissions, then Apps and services, choose Vidljivo and remove its access. Mapped TikTok report data stops being available for ordinary analysis after 30 days from its original supporting read, with routine cleanup of expired fields and the tokens we hold are deleted from our systems within 24 hours of disconnecting (section 7). Questions: hello@vidljivo.com, or see the data deletion page (vidljivo.com/legal/data-deletion). 7. How long we keep data • Data: Account and work data: account, brands, content and media, scans and findings, agent conversations; How long we keep it: While the account is open. After you close it and choose deletion, deletion is carried out step by step and is due within 30 days of the moment it starts; an export stays open for 30 days first; Why: To provide the service and to delete what you ask us to delete • Data: Mapped non-daily reports from connected Google, Meta and TikTok services; How long we keep it: Up to 30 days from the original supporting read. After 24 hours the read is marked older; its date and measurement period remain distinct. Routine cleanup removes expired report fields; this is not a promise of instantaneous physical erasure; Why: To provide current and dated decision context without treating an older observation as a new read. Current account and action checks may require a refresh. Connection controls, saved content, approvals and outcome/security records follow their separate purposes and periods • Data: Qualified dated daily figures: Search Console and GA4 aggregates; daily spend, impressions, clicks, link clicks and reach of a Meta campaign created through Vidljivo, where offered; daily spend, impressions and clicks of a selected TikTok advertiser account; daily impressions, clicks and, only where that read collected it, cost of a Google Ads campaign Vidljivo created for you; How long we keep it: Up to 13 months by the measured day, where supported, in the currency and time zone the source reports; the measured day and the date it was read stay distinct. No complete history is promised: days may be missing or later restated by the platform. Cost is kept only from reads that collected it and is not added to earlier records. The Ads daily sources listed here do not store ad-conversion, attribution or return-on-ad-spend histories; supported GA4 event and commerce aggregates follow their own definitions; Why: To show results over time, only where dated figures and supporting evidence actually exist • Data: Current inputs for automatic Search Console opportunity triggers; How long we keep it: The latest input must have been read within 30 days. Qualified dated historical context follows its separate report and aggregate retention periods; Why: To avoid starting a new opportunity from old current-state inputs while preserving useful historical comparisons • Data: Your conversations with the agents in the team chat; How long we keep it: 35 days; Why: To continue the conversation and to show what was answered • Data: Notes the agents keep from the team chat, when chat memory is on; How long we keep it: Until you remove the note in Settings, or close the brand or the account; Why: So that the team remembers what you told it • Data: The record of your ads’ results (section 6); How long we keep it: Daily figures 30 days, then only totals; each ad up to 13 months after its last measured day. Erased at once when you stop sharing platform numbers, 30 days after you disconnect Meta Ads or Google Ads unless you connect again, and with the brand or the account; Why: So that Kora plans better ads from real results • Data: Usage records of the AI team (what each call used; no text); How long we keep it: 35 days; Why: To enforce limits and to detect abuse • Data: E-mails you send us and our replies, as the threads our mail system keeps; How long we keep it: 90 days after the last activity in the thread. Threads that are pending, held, failed or handled by a person, or that belong to an open task, are kept while they are open and at most 1 year. Threads about a privacy request, a complaint, a withdrawal, a take-down request or another legal matter are not cut by that limit while the matter is open; once it is closed, the record of it is kept as the complaint records row of this table says. The threads of a closed account are deleted at once; Why: To handle your request and to show what was answered • Data: The record our mail system keeps of each message sent to us (the intake record); How long we keep it: Its subject, text, attachment names, internal notes, the sender’s name, the sender’s e-mail address and its domain, the domains that sent and signed the message, and every other address in it are erased at the latest 90 days after we received the message (1 year if it is still being handled or is held for a person). What is left of the record after that holds no name, address or domain: only the class and status we gave the message, how it was handled, its size, its dates, its language and technical numbers. It is deleted at the latest 1 year after we received the message. While it still holds the address it is also deleted when the account that uses that address is closed. It is also deleted when you ask us to erase your mail: we erase the mail that came from the address you write to us from, except for a complaint, a data-protection request or another legal matter, which we keep as a record of its own (see complaint records); Why: To show how each message was handled, and to find its record again when an account is closed or when you ask us to erase your mail • Data: The original messages in our mail provider’s mailbox; How long we keep it: The original messages stay in our mail provider’s mailbox (Zoho Mail, EU) until we delete them. We delete them on request, or under a retention rule once one is set, and we promise no date or period for that deletion until such a rule exists. Why: To keep the original of a message you sent us • Data: Invoices and tax and accounting records; How long we keep it: 6 years after you close the account; Why: Spanish commercial and tax law • Data: The record of which legal documents you accepted and which choices you made; How long we keep it: 6 years after you close the account; Why: To defend legal claims • Data: Payment evidence: billing country, card country, IP country, tax-ID status, invoice reference and rate; How long we keep it: 10 years; Why: Tax law and proof of the place where the service is supplied • Data: Referral credit records applied to invoices; How long we keep it: 6 years after you close the account; Why: They follow the payment ledger • Data: Refunds, disputes and the audit entries of our billing system; internal handling tasks; How long we keep it: 6 years from the event; internal handling tasks 1 year after they are resolved; Why: Commercial records and defence of legal claims • Data: Complaint records; How long we keep it: 6 years from closure; after 12 months only the provider can read them; Why: To defend legal claims • Data: The record of a request, sent to us by Facebook, to delete your data: the date, how many links to your Facebook account were removed and a one-way fingerprint of your Facebook account number (no name, no e-mail address); How long we keep it: 6 years; Why: To show that we did what Facebook asked, and to answer its status link • Data: Security and audit logs; How long we keep it: 90 days; Why: Security of the service • Data: A salted hash of the network prefix of your connection (the IPv4 address, or the IPv6 /64 and its /48 group), per account; never the address itself; How long we keep it: Refreshed every 30 days while the account is open; erased with the account. Copies of the hash on the usage records of the AI team are kept as long as those records are (35 days) can remain after account closure for their original usage/security lifetime; closure does not start a new period; Why: To limit abuse of the allowance that all Free accounts share; used by us only, no recipient • Data: Credentials of connected accounts (tokens); How long we keep it: While the account is connected; deleted from our systems within 24 hours of disconnecting or closing; Why: Needed to act on the account you connected • Data: Feedback and ratings you give about the service; How long we keep it: 12 months; Why: To improve the service • Data: Accounts whose e-mail address was never confirmed; How long we keep it: Closed and deleted automatically 7 days after sign-up, by a check that runs every hour; only the record of the acceptance made at sign-up is kept, as the row on acceptance records says; Why: To prevent abuse and not to keep the data of people who never finished signing up • Data: Sign-in sessions and one-time links (e-mail confirmation, password reset, steps of connecting an account); How long we keep it: Valid only until they expire; an automatic check that runs every hour then deletes them, within 24 hours after they expire at the latest; Why: Security of the service • Data: Backups; How long we keep it: Overwritten as they expire, within 30 days; a restored backup is erased again automatically; Why: Resilience; deleted data leaves the backups within that period • Data: The stub of a closed account (the account number and the plan: no name, no e-mail address, no content); How long we keep it: No end date; Why: So that the number is not reused and a deletion can be shown • Data: A count-only record that a deletion was carried out (dates, numbers and the account number, no content); How long we keep it: 60 days after the last backup that held the data; Why: To show that we did what we said Legal records are not deleted on request before the end of their period (GDPR art. 17(3)(b) and (e)). 8. Your rights Under the GDPR you may request access, rectification, erasure, restriction and portability, object to processing based on legitimate interest, and withdraw consent at any time. Write to hello@vidljivo.com; we may verify your identity, and we answer within one month of receiving your request. You can complain to the Spanish data-protection authority (Agencia Española de Protección de Datos, aepd.es) or to the authority of your country of residence. Account and data deletion is described on the data deletion page (vidljivo.com/legal/data-deletion); it does not reach the records that section 7 says we must or may keep. A copy of your data contains what you entered or received in the service, including the e-mail addresses of other people that you typed, and the e-mails you sent us. 9. Cookies, local storage and measurement We use necessary first-party cookies and browser storage for sign-in, devices, security and the service you request. The optional referral cookie is separate. Optional Google Analytics measurement on our public marketing website is also separate: it starts only if you select Allow analytics. Rejecting it does not prevent you from using the website, registering, using the app or connecting a business. An account or connected-account permission is not consent to this visitor measurement, and this visitor choice is not permission to share your connected business data with AI agents. Cookies • Cookie: __Host-v_session; What it does: Keeps you signed in; Duration: 7 days, or until you sign out • Cookie: v_device; What it does: Recognises a browser you have already signed in with, so that failed sign-in attempts from elsewhere cannot lock you out; its content is sealed and only our service can read it, and it is set only after a successful sign-in; Duration: 180 days • Cookie: v_oauth; What it does: Protects the step of signing in with Google or Facebook; Duration: 10 minutes • Cookie: v_social_; What it does: Protects the step of connecting a social account (the platform name follows: meta, instagram or tiktok); Duration: 10 minutes • Cookie: v_tiktok_business_oauth; What it does: Protects the step of connecting TikTok Ads to your signed-in brand; Duration: 10 minutes, or until the connection returns • Cookie: vidljivo_ref50; What it does: Remembers the referral link you arrived through, to apply the referral discount when you create an account; set only if you agree to it; Duration: At most 30 days When you link a website to its owner, the necessary __Host-vidljivo_owner_link cookie protects that browser step for 5 minutes and is cleared on a completed return. It is not an analytics cookie. After your analytics opt-in, Google Analytics can set the first-party cookies _ga, which distinguishes browsers, and _ga_G1WB5NN3J9, which maintains session state. We configure these analytics cookies to expire after 30 days, measured from their initial creation rather than extending the expiry at each visit. Browser restrictions may shorten that period. These cookies are optional and are not used by us for advertising or advertising personalisation. Our code does not load the Google tag or send consent pings before you opt in. These optional analytics cookies are separate from the necessary service cookies described above. Browser storage • Key: vidljivo.ad-launch:*; Where: session storage; What it keeps: The progress of launching an advertisement (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.ad-plan:*; Where: session storage; What it keeps: The progress of preparing an advertisement (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.ads-source-budget:*; Where: session storage; What it keeps: The budget you are changing in an advertising proposal (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.ads-source-closure:*; Where: session storage; What it keeps: A request you confirmed to close a paused Google ad setup and withdraw its earlier activation approval, kept so that only that same request is checked or finished (followed by your account and brand identifiers and website address); Duration: Until the tab is closed or the closure is confirmed • Key: vidljivo.ads-source-draft:*; Where: session storage; What it keeps: The draft of an advertising proposal you are editing (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.ads-source-schedule:*; Where: session storage; What it keeps: The schedule you accepted for a paused advertisement (the dates read from the ad platform), with a copy of the reviewed advertisement and setup, kept so that only that same request is checked (followed by your account and brand identifiers and website address); Duration: Until the tab is closed or the request has a final answer • Key: vidljivo.ads-source:*; Where: session storage; What it keeps: The progress of an advertising proposal (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.ads-strategy-later:*; Where: session storage; What it keeps: That you chose to decide later on a set of advertising strategies: a fingerprint of that set, which keeps it hidden in this tab until you show it again or a new set appears (followed by your account and brand identifiers and website address); Duration: Until the tab is closed or you show the strategies again • Key: vidljivo.ads-strategy:*; Where: session storage; What it keeps: An advertising strategy step you started (saving a spending limit, asking for strategies or choosing one), kept until the server answers so that only that same request is retried (followed by your account and brand identifiers and website address); Duration: Until the tab is closed or the server answers • Key: vidljivo.agent-decision-write:*; Where: local storage; What it keeps: An answer you sent to an agent proposal (accept, dismiss or apply) whose result is not yet known: the action and identifiers only, never the proposal text, kept to block a conflicting answer (followed by your account, brand and proposal identifiers); Duration: Until the proposal shows a state that answer can no longer change, or until you delete it • Key: vidljivo.analytics.public.v1; Where: local storage; What it keeps: Your separate optional public-site Google Analytics allow/deny choice, decision/expiry times and exact Privacy notice version/fingerprint (local storage in the real component); Duration: 30 days from the decision; a changed notice, expiry or withdrawal disables collection • Key: vidljivo.google-connect.*; Where: session storage; What it keeps: The progress of connecting a Google account (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.google-stop:*; Where: local storage; What it keeps: A confirmed request to pause this Google ad's own campaign, including the reviewed stop, its request identifier and your account, brand, website and ad identifiers. Kept so a later check uses the same request; a lost reply or error does not discard it; Duration: Until the same request returns a verified or closed-without-dispatch result, or until you delete it • Key: vidljivo.guide.v1.*; Where: local storage; What it keeps: Whether you have seen the workspace guide and at which step (followed by your account number); Duration: Until you delete it • Key: vidljivo.language; Where: local storage; What it keeps: Your language choice; Duration: Until you delete it • Key: vidljivo.layout.*; Where: local storage; What it keeps: Whether a page shows cards or a list (followed by the page name: activity, posts or library); Duration: Until you delete it • Key: vidljivo.meta-continuation:*; Where: local storage; What it keeps: A request you confirmed to finish creating a paused Meta ad setup, with a copy of the reviewed steps, kept so that only that same request is sent or checked (followed by your account and brand identifiers, website address and the identifiers of that ad); Duration: Until the setup is completed, stopped or not admitted, or until you delete it • Key: vidljivo.meta-quarantined-continuation:*; Where: local storage; What it keeps: A request you confirmed to complete a paused Meta ad setup held for extra review, with a copy of the reviewed setup, your confirmations and the capacity you act in (owner or agent for the owner), kept so that only that same request is sent or checked (followed by your account and brand identifiers, website address and the identifiers of that ad); Duration: Until the setup is confirmed, held or not admitted, or until you delete it • Key: vidljivo.meta-stop:*; Where: local storage; What it keeps: A stop you confirmed for a Meta ad (pause its campaign or retire its paused setup), with a copy of the reviewed stop, kept so that a later check reuses that same request instead of creating a new one (followed by your account and brand identifiers, website address and the identifiers of that ad); Duration: Until the stop has a final result or is refused before it starts, or until you delete it • Key: vidljivo.profile-language; Where: session storage; What it keeps: The language of your profile, used as a fallback in this tab only; Duration: Until the tab is closed • Key: vidljivo.provider-intent:*; Where: session storage; What it keeps: A change to a connected account you are reviewing (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.social-connect.*; Where: session storage; What it keeps: The progress of connecting a social account (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.support-intent.v1:*; Where: session storage; What it keeps: A support request you are preparing (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo.tiktok-ads-connect.*; Where: session storage; What it keeps: The progress of connecting a TikTok Ads account (followed by your account identifier); Duration: Until the tab is closed • Key: vidljivo.website-oauth.*; Where: local storage; What it keeps: Binds the same website connection attempt to its provider, account, brand, site and expiry (session storage in the actual website connector); Duration: Until the tab session ends or the original flow is completed or cancelled; expired flows cannot authorize a connection • Key: vidljivo.website.*; Where: local storage; What it keeps: The website you last chose (followed by your account number); Duration: Until you delete it • Key: vidljivo.wordpress-connect.*; Where: session storage; What it keeps: The progress of connecting a WordPress site (followed by identifiers of that step); Duration: Until the tab is closed • Key: vidljivo:chunk-reload-at; Where: session storage; What it keeps: When the app last reloaded itself after an update, so that it reloads at most once in a short time; Duration: Until the tab is closed • Key: vidljivo:kora-position:ui:*; Where: local storage; What it keeps: Where you placed the floating assistant button (followed by your account number); Duration: Until you delete it Your separate analytics choice is stored under vidljivo.analytics.public.v1 for 30 days, together with its decision and expiry times and the version and fingerprint of this notice. A changed notice, invalid preference or expired choice requires a new decision before measurement starts. Optional public-website analytics When you consent, our public marketing pages send Google Analytics the canonical public page address without query strings or raw fragments, the fixed page title “Vidljivo” and named marketing-section events selected from a fixed list. We supply an empty referrer. We do not supply your account or brand identifier, email, form text, conversation, sign-in or connection tokens, or private app content. Measurement is disabled on login, registration, connection, billing and private workspace screens; it depends on both a public address and the actual public-page renderer, not just on being signed out. Google’s tag also receives browser and device information, network information including the IP address used for the request, and pseudonymous browser/session identifiers. This optional measurement is therefore not anonymous or identifier-free. We disable Google Signals, advertising storage and ad-personalisation signals. We do not claim that this stage measures completed registrations, sales or advertising attribution. Google Analytics is a separate recipient from the AI model hosts listed above. Requests go directly from your browser to Google and processing may occur outside the European Economic Area, including in the United States, under the applicable Google terms. See Google’s privacy information (policies.google.com/privacy) and Analytics data safeguards (support.google.com/analytics/answer/6004245). We do not claim that all Analytics processing stays in Europe or that the service’s other processor agreements automatically cover this separate purpose. For our public-site Analytics property, the current event-data setting is 2 months and the user-data setting is 14 months, with user retention reset on new activity. Those settings do not limit standard aggregated reports in the same way. They are separate from the app’s retention of data retrieved from a customer’s connected property and from the expiry of cookies in your browser. You can withdraw this consent in one click through Turn analytics off in the public website’s footer. This stops future measurement and attempts to remove this property’s analytics cookies; it does not recall events Google has already received. If browser storage or the current notice cannot be verified, measurement stays off. Your rights in section 8 also apply. Our identifier-free counters Separately, when our own service measurement is enabled, our server keeps daily counters for page groups and a few named steps without a visitor or account identifier. These counters do not use a measurement cookie, browser identifier or third-party measurement service. They do not record keystrokes or page text. Anyone can submit requests to those endpoints, so they are indications rather than exact visitor figures, and they are retained for 13 months. The optional Google Analytics collection above is a different system and does use pseudonymous identifiers. 10. Security We protect data with encryption in transit, encryption of stored credentials, access controls and organisational measures, and we limit access to what each task needs. We keep credentials out of logs and error reports. 11. Sending a conversation to customer support In the team chat, when the agents conclude that they cannot help, or when your message is a matter that only the provider handles (for example a legal claim, a data-protection request or a complaint), the chat offers a button to send the conversation to customer support. Nothing is sent unless you click it, and only that conversation is sent. The messages of that conversation (yours and the agent’s, shortened if they are long), the name of the agent, the language and the e-mail address of your account are sent to customer support, where the provider reads them and follows up with you by e-mail. The record that a conversation was sent is kept like the support messages of section 7. 12. Children Vidljivo is not for people under 18, and we do not knowingly process their data. 13. Changes to this notice We update this notice when our processing changes and post the new version here. For a material change we tell you in the app or by e-mail 30 days before it applies.