Privacy policy

Beseen measures whether Google ranks you and whether AI assistants name you, then writes the fix. That work runs on public web pages and on answers from AI assistants, so the personal data involved is small and mostly about you as a customer rather than about your visitors.

• To use Beseen you give us a domain and an email address. Billing details go to our payment processor, not to us.
• We crawl the public pages of the site you added, the way a search engine does. We never get a login, a CMS connection, or write access to your site.
• We generate prompts about your category and send them to ChatGPT, Claude and Perplexity to see who gets named. Those prompts are about products, not people.
• We store the results: your positions, the prompts you lose, the pages we crawled, and the copy we wrote for you.
• We do not sell personal data, and there are no advertising trackers on this site.

Beseen is operated by Sparkeit SRL, registered in Romania. For the data described here we are the controller, which means we decide what is collected and why.

Questions about this policy, or any request about your own data, go to support@beseen.so. We have not appointed a formal data protection officer; that email reaches the people responsible for this instead.

Your account

An email address, a display name if you give one, and either a password hash or, if you sign in with Google, your Google account identifier. Also your plan, and the sites and prompts you have chosen to track.

Billing

Payments run through Stripe. Card numbers are entered on their side and never reach our servers. We keep what we need to run the subscription and the accounts: your plan, its status, invoices, the billing name and country, and the last four digits of the card.

The site you asked us to track

The domain, and the content of the public pages we crawl: titles, meta descriptions, headings, body copy, internal links and the HTTP behaviour of those pages. If your public pages carry personal data, an author bio, a team page, a phone number in a footer, then that text sits in our crawl records too. We do not go looking for it and we do not use it for anything except writing suggestions for that page.

What we measure

Keyword positions and history, the competitor domains we discovered, the prompt sets we generated, the answers each assistant gave, which brands were named and cited in them, and how readable your domain is to machines. This is data about websites. It only becomes personal data through its link to your account.

Free public scans

The free scan takes no signup. We record the domain entered, the report it produced, and the requesting IP address so we can rate limit it. More on what happens to that report below.

Technical logs

IP address, browser user agent, timestamps, pages requested and errors. Standard server logs, kept for security and debugging.

• To run the product you signed up for: discovery, crawling, prompt checks, the list of prompts you lose, rewrites and the weekly re-check. Scheduled checks run whether or not you log in, because that is what makes the re-check a before and after.
• To bill you and keep the records that go with that.
• To email you: your report, a re-check result, a billing notice, a password reset.
• To keep the service up and stop abuse: rate limiting, fraud checks, debugging.
• To improve how well we generate prompt sets and suggestions, by looking at which changes actually moved a citation. That analysis runs on site and ranking data across customers, in aggregate, and does not need your account details.

The legal bases

• Performing our contract with you: running the scans, storing the results, billing.
• Our legitimate interests: security, abuse prevention, and improving the product in the aggregated way described above.
• Legal obligation: keeping invoices and tax records.
• Consent: marketing email, which you opt into and can leave at any time. Nothing about the running of your account depends on it.

Account records, crawl results, prompt sets, assistant answers and generated copy live in our database, hosted by Neon in the United States. The application itself runs on Vercel. Backups run on the same infrastructure as the database and are encrypted at rest. Access inside the company is limited to the people who need it to run or support the service.

Sub-processors

These are the outside services that handle data on our behalf, and what each one gets.

• DataForSEO is our data layer. It receives your domain and keywords, runs the page crawl, returns ranked keywords, competitors and SERP data, and routes our prompt checks to the AI assistants.
• OpenAI, Anthropic and Perplexity receive the prompts we generate and return the answers plus their citations. We reach them through DataForSEO's AI Optimization API rather than directly, so their terms and DataForSEO's both apply to those requests.
• Anthropic also does the synthesis and the writing, so text from your crawled pages passes through Claude when we produce a diagnosis or a rewrite.
• is-agentic.com receives your domain once at onboarding and scores how readable it is to machines. It is a free public API and it publishes stable report URLs for the domains it scores.
• Stripe handles payments and holds your billing name, address, card details and subscription history.
• Vercel hosts the application and the server logs.
• Neon hosts the database and its backups: account records, crawl results, prompt sets, assistant answers and session records.
• Resend sends transactional email and therefore holds your email address and the contents of those messages.
• Google sees your email address and basic profile information if you choose to sign in with Google. We do not use any other third party for authentication.
• Vemetric sees usage events from the site and the dashboard, which include IP addresses and browser details.

Where the data goes

Sparkeit SRL is registered in the EU, but the services above run in the United States. If you are in the EU or UK, this means your data is transferred outside it. That transfer relies on standard contractual clauses, or the EU-US Data Privacy Framework where a provider is certified. Ask us at support@beseen.so and we will tell you which service is involved in a given piece of processing.

The crawl reads public pages only. There is no login to your CMS, no OAuth against your site, and no write access of any kind. Beseen hands you copy to apply. It never changes a page on your behalf.

• A discovery run covers up to roughly 100 pages of the domain you added.
• Rewrite suggestions are capped at 20 pages per run, and there is no bulk apply-all, because both of those are how an automated tool quietly wrecks a site that already ranks.
• The fetch is made by DataForSEO on our behalf, so it arrives with their crawler identity rather than ours.
• We store what the crawl returns for as long as you track that site. Ask us and we will delete a site and everything attached to it.
• We crawl the site you added. Competitor analysis runs on index data and assistant answers, not on a crawl of anyone else's pages.

For each tracked site we generate 20 to 30 prompts a buyer in your category would actually ask, then run them across ChatGPT, Claude and Perplexity on your plan's cadence and record which brands each answer names and cites. Those answers are stored so we can show you what changed between one check and the next.

Prompts are about products and categories, not about people, and we generate them from keyword and competitor data rather than from anything personal. If you edit a prompt yourself, keep personal data out of it: whatever is in a prompt leaves our systems and reaches three separate providers.

The free scan runs five prompts on one engine and produces a report at a shareable URL. Shareable means exactly that: anyone with the link can open it and see the domain scanned, the prompts we ran and which competitors were named. It is meant to be passed around, so do not run it on something you would rather keep quiet.

We keep the requesting IP address for rate limiting for 30 days, and the report itself for 90 days. Email support@beseen.so with the link and we will take a report down.

• Account data: while your account is open, then deleted within 30 days of you closing it.
• Tracked site data, including crawls, prompt sets and assistant answers: while you track that site, then deleted within 30 days of you removing it.
• Invoices and billing records: as long as tax law in Romania requires, which outlasts your account.
• Server logs: 30 days.
• Free scan reports: 90 days.
• Aggregated outcome data, meaning what kind of change moved a citation in what kind of category: kept indefinitely, with no link back to you or your account.

Depending on where you live you can ask us for a copy of your data, ask us to correct it, ask us to delete it, ask for it in a portable format, object to processing we base on legitimate interests, or withdraw consent you have given.

Email support@beseen.so and we will answer within a month. You can also delete a tracked site, or your whole account, from the dashboard without asking us. If you think we have handled your data badly, you can complain to ANSPDCP, Romania's data protection authority or to the supervisory authority where you live or work.

We set a cookie to keep you signed in, and one to remember cookie choices. Vemetric runs without setting a cookie of its own. There are no advertising cookies and no third-party ad networks on this site.

Beseen is a tool for businesses and it is not aimed at children. Do not use it if you are under 16. We do not knowingly collect data about children, and if we learn we have, we delete it.

This version is dated 2 September 2026. If we change something that affects what we collect or who touches it, we will email account holders before it takes effect rather than quietly editing this page.

Questions go to support@beseen.so. The rules for using the product itself are in the terms of service.