Privacy Policy

Semantic Checker operates this service and is responsible for the data described here. This policy covers the Semantic Checker web application and its marketing pages. Questions and deletion requests go to thirdweb30@gmail.com.

• —Your Google account details. Signing in with Google is the only way in. Google returns your name, email address, profile picture URL, and an account identifier, and we store those along with the OAuth tokens needed to keep you signed in. We never receive or store a password — there is no password on this service to steal.
• —The content you analyze. Your target query, the optional title, and the full article text are stored with your account so reports stay available at their own URLs. Text you bring in through Import URL is stored the same way as text you paste.
• —Reports. Each analysis result — scores, section verdicts, evidence, and priority fixes — is stored alongside the content it describes.
• —Your credit ledger. An append-only record of every credit granted, spent, and refunded. It is never edited or deleted in place, because it is what makes your balance auditable.
• —Billing pointers. A customer identifier and subscription state from Stripe. Card numbers never reach our servers — Stripe collects and holds them.
• —Session and operational records. Session tokens with the IP address and user agent that created them, plus ordinary server logs used to keep the service running and to enforce rate limits. To avoid paying for duplicate SERP requests, we keep a keyed one-way digest of the target query and the public result snapshot for up to six hours. The cache does not contain the query itself, your title, or your article text.
• —Welcome-credit abuse controls. Before Google sign-in we set a signed first-party cookie containing a random device identifier. For the promotional-credit decision, we store keyed one-way digests of that identifier, your normalized IP address, and your email identity, together with the decision and time. We do not store a new copy of your raw IP address in this anti-abuse record.
• —Product analytics. Google Analytics records which pages are opened and which interface buttons are used. Microsoft Clarity records interaction and diagnostic data, including session playback and heatmaps. The signed-in application content area is explicitly masked from Clarity. Both may receive standard device, browser, language, and referring-site details, but neither is configured to receive your content.

• —We do not sell or rent your data, and we do not share it for advertising.
• —No ad tech. No advertising or ad-targeting scripts are bundled in this application. Clarity is used for behavioral analytics, not for advertising by us, and the advertising side of Google Analytics — Google signals, ads personalization, and the third-party advertising cookies that come with them — is left switched off.
• —Cookies and local storage: the cookie that keeps you signed in, the signed first-party device cookie used to limit promotional welcome credits (kept for up to 400 days unless your browser removes it sooner), your light/dark theme preference stored in your own browser, the first-party _gacookies Google Analytics uses to tell a returning visit from a new one, and Clarity's first-party _clck and _clsk cookies, which preserve its pseudonymous user ID and connect page views into a session. Depending on the browser and Clarity configuration, Microsoft may also set third-party cookies such as CLID, MUID, ANONCHK, MR, and SM.
• —We do not use your articles to train machine-learning models.

We run Google Analytics 4 and Microsoft Clarityon the marketing pages and inside the signed-in app. Google Analytics shows which pages people arrive on and which interface buttons are used. Clarity captures interaction, diagnostic, and session-playback data to help us understand usability. The signed-in application's main content area is marked for Clarity masking, and input fields are masked by Clarity. We do not intentionally send article text or target queries as analytics parameters. Events report the page path and a short, fixed interface label such as a button's accessible name; after a successful Google sign-in, one Google Analytics event also includes the account name and email address you used to sign in. Neither service is configured to receive article or query text. Both may receive ordinary device, browser, language, and referring-site details. Google infers an approximate location — roughly city level — from your IP address; the address itself is not stored in the reports we see.

Analytics never sees your content. Your target query, your article text, and your report contents are sent to our own server in requests the analytics tag cannot read, and a report address contains nothing but a random identifier. We do not attach your account to ordinary analytics events. The post-login event is the exception: it includes your Google account name and email so we can distinguish a real sign-in from anonymous traffic.

To opt out, use your browser's cookie controls or any content blocker. Google Analytics users can also use Google's official opt-out browser add-on. Everything on this site works exactly the same with analytics blocked.

Import URL makes our server request the page you name and extract its text. That request comes from our infrastructure, not your browser, so the destination site sees us rather than you. We only follow public HTTP and HTTPS addresses, refuse private and internal network destinations, and stop after a few redirects. We store the extracted text, not a copy of the page. Importing is free and does not spend a credit — only running the analysis does.

We rely on a small set of processors, each doing one job: Google for sign-in and Google Analytics, Microsoft for Clarity, our database and hosting providers for running the service, Stripe for payments, and Resend for the handful of transactional emails we send. Upstash enforces shared rate limits and keeps the short-lived SERP snapshot cache described above. DataForSEO receives the target query—not the draft—to return a live US-English Google result set. When enabled, Google Cloud Natural Language receives the article text to identify entities and their prominence. Where a self-hosted deployment substitutes different providers, this list changes with it.

Google Cloud Natural Language does not provide a Google Search ranking score. We show its entity output as supporting evidence and keep the editorial decision with you. Our server also fetches public competitor pages named in the result set and compares their extracted topic language with the draft.

Account data, stored articles, and reports are kept until you delete them or delete your account. Deleting your account from Settings removes your profile, your Google connection, every stored analysis and report, and your ledger history, and cancels any active subscription first. Records Stripe is required to retain for tax and accounting purposes stay with Stripe. Backups roll off on their own schedule.

The short-lived SERP cache expires automatically after six hours. It stores public result metadata behind a keyed one-way digest, not the target query, title, article text, or account identifier.

One set of records does outlive the account, and we would rather name it than bury it: keyed one-way cryptographic digests of the email identity and, when available, the signed device identifier and normalized IP address used for the welcome-credit decision. We keep the decision and its date indefinitely as an anti-abuse record to enforce one grant per email and device; an IP digest only affects eligibility for the preceding 30 days. The digests cannot be turned back into the underlying values by someone holding only the table, and they are not linked to your profile, articles or reports.

Analytics events and Clarity session data are the exception, and we would rather say so than leave it implied: they live in Google Analytics and Microsoft Clarity. Google Analytics uses its property retention settings; Clarity recordings are retained for up to 30 days under its current standard session-recording policy. These records are not removed when you delete your account — there is no link between the analytics records and your account to follow.

You can view and edit your name and email in Settings, delete individual reports, or delete the whole account at any time. Depending on where you live you may also have the right to request a copy of your data, ask for corrections, or object to particular processing — write to thirdweb30@gmail.com and we will respond.

Account deletion is confirmed by an emailed link, because with Google-only sign-in there is no password to re-enter.

Semantic Checker is a tool for people publishing content professionally and is not directed at children. We do not knowingly collect data from anyone under 16.

When this policy changes we update the date at the top of this page. Material changes to what we collect or who processes it will be announced by email before they take effect.