Privacy Policy

Last updated: 6 October 2026

This privacy policy explains how DATA HIT Ltd ("DATA HIT", "we", "us", "our") handles your information when you use the tools hosted at tools.datahit.co. The tools fall into two categories with different data-handling models — both are covered in this single policy. Read alongside the DATA HIT main privacy policy, which covers general website usage.

DATA HIT Ltd is a company registered in England and Wales, with its registered address at Department, 4 The Boulevard, Leeds, LS10 1PZ. For any privacy-related enquiries, please contact us at hello@datahit.co.

1. The tools at a glance

We currently offer the following tools at tools.datahit.co:

• Seasonal Spiral & Timing Grid (free tier) — connect to your own Google Analytics 4 (GA4) and Google Search Console (GSC) accounts via OAuth 2.0 to generate browser-side visualisations. No data leaves your browser other than calls direct to Google's APIs.
• Search Seasonality (free tier; a free account for keyword sets, Pro for saved sets) — submit a keyword or a keyword set and a country; we fetch monthly Google search volume history via a third-party SEO data provider. Server-side processing is involved; details in section 3 below.
• AI Answer Monitor (free tier; Pro tier for topic scans and prompt checks) — the free AI-answer exposure report takes your domain and a country and asks a third-party SEO data provider which search terms your site ranks for, and which of those Google answers with an AI Overview; we visualise the share of your search demand those answers cover. Pro adds topic scans (submit a keyword and an AI engine; we retrieve the brands, fan-out queries and cited sources that engine surfaced) and prompt checks (submit one exact prompt and we run it live). Server-side processing is involved; details in section 4 below.
• AI Answer Stability (free account; more checks on Pro) — asks the AI engines the same question several times and counts how often each names your brand. See section 4a.
• Seasonal Spiral (Data Studio) — a community visualisation that runs entirely inside Data Studio's sandboxed iframe. No data reaches DATA HIT servers.
• DATA HIT Tools Pro (paid tier) — optional account-based add-on that layers persistent features on top of the free tools: a larger monthly allowance of AI credits, tracked keywords with weekly SERP snapshots, watched AI Answer Monitor scans (weekly re-runs are paused at present), saved view configurations, cross-tool annotations, AI overviews on the charts, and a Monday-morning email digest. Requires an email address and a Stripe-managed subscription. Server-side state is involved; details in section 5 below.

2. OAuth-based tools (Seasonal Spiral, Timing Grid)

When you connect your Google account to one of the OAuth-based tools, we request the following read-only scopes:

• analytics.readonly — used to list your GA4 properties (via the Admin API) and retrieve daily aggregate report data (sessions, transactions, revenue, event counts via the Data API).
• webmasters.readonly — used to list your Search Console sites and retrieve daily aggregate search performance data (clicks, impressions via the Search Analytics query endpoint).

Both scopes are read-only. We do not request any permissions to modify, delete, or write data to your Google accounts.

Spiral and Timing Grid each offer two ways to connect. The default is browser-only and stores nothing on our servers. The optional persistent connection stores a refresh token on our servers so the connection survives across visits and so Pro features that run on a schedule (e.g. the Monday digest) can fetch on your behalf.

Default: in-browser connection.

• No server-side storage. Your Google data is never transmitted to, processed by, or stored on any DATA HIT server.
• No database. We do not maintain any database of user data, analytics data, or access tokens for this path.
• Browser memory only. All data retrieved from Google's APIs exists only in your browser's memory for the duration of your session. It is discarded when you close the page, refresh the browser, or disconnect.
• No refresh tokens. The in-browser flow (OAuth 2.0 implicit flow / Google Identity Services token client) does not issue or store refresh tokens. Access tokens are short-lived (~60 minutes) and held in browser memory only.

Optional: persistent server-side connection. When you click "Stay connected next time" beside the Connect Google button, or any time a Pro feature needs to query Google on your behalf when you are not actively at the page (e.g. the Monday digest cron), the tool initiates the OAuth 2.0 authorisation-code flow with offline access instead. This issues a refresh token that we encrypt and store server-side so we can mint short-lived access tokens for future requests. Specifics:

• What is stored. The refresh token (encrypted at rest), the Google account email, the selected GA4 property or Search Console site, and timestamps. One row per user per provider in D1 (ga4_connections, gsc_connections).
• What is not stored. Access tokens (only ever held in memory for the duration of one API request) and your analytics data (re-fetched on demand, never persisted to a DATA HIT database).
• Scopes. analytics.readonly for GA4, webmasters.readonly for Search Console. Read-only. Same scopes as the in-browser path.
• Revocation. Click "Disconnect" in the tool, or visit your Google Account permissions page. Either path soft-deletes the row immediately and we no longer receive access. The encrypted refresh token is purged from the database on hard delete at account closure.

We do not use your Google user data for any purpose other than providing the visualisation and the Pro features you have enabled. Your data is never used for advertising, marketing, profiling, AI model training, or any purpose unrelated to the tool's core functionality.

To revoke access at any time, click "Disconnect" in the tool or visit your Google Account permissions page.

3. Search Seasonality (server-side data flow)

Search Seasonality runs partly on our servers because it looks up monthly search volume from a third-party SEO data provider. This section explains exactly what is sent where.

What you provide:

• A search keyword (e.g. "christmas jumper"), or a list of keywords for a keyword set, entered by you, uploaded as a CSV, or imported from your Search Console.
• A country selection (one of the eleven countries the tool supports).

Looking up a single keyword needs no account and we do not ask for any personal information. Keyword sets need a free account.

What we process and keep:

• Your IP address — used by Cloudflare's edge to apply per-IP rate limits (10 scans per minute). The IP is not logged to our database or shared with third parties; it is consumed transiently by the rate-limit binding and then discarded.
• Search volume cache — each keyword's monthly search volume is stored in Cloudflare Workers KV, per keyword and country, until the next month's data is due (usually a few weeks), so repeat lookups do not re-query our data provider. The cache holds the keyword, country and monthly volumes only; it does not record who looked the keyword up.
• Keyword-set scans (signed in) — for each set you scan we record your account id, the time, the country, how many keywords the set had and the provider's cost, to apply your plan's daily limit. We do not record the keywords themselves in this log.
• Saved sets (Pro) — if you save a keyword set we store its name, country and keyword list with your account until you delete it or close your account. If you are on Pro Agency you can file a set under one of your client workspaces.
• Weekly email lines (Pro) — so the Monday email mentions each saved keyword once when its demand starts increasing and once in its peak month, we record which keyword, country and month we have already written about, linked to your account.
• Share links — links created with the previous version of this tool (SERP Seasonality Map) keep working for up to 30 days and now open the keyword in Search Seasonality. Each maps a random 8-character slug to a keyword and country in Cloudflare Workers KV.

Search Console import: on a free account, the import signs you in to Google in your browser with read-only Search Console access, and your top queries go straight from Google to the page; they are not sent to or stored by DATA HIT. On Pro with a saved Search Console connection, our server fetches the property's top queries with that connection and passes them to your browser; we do not store them. Either way, nothing is sent to our data provider until you choose to scan.

Third parties that receive your scan data:

• DataForSEO (privacy policy) — receives the keywords you scan and the country code, to look up their monthly Google search volume history. For keywords with a short history we also ask its Google Ads search-volume endpoint for up to four years of monthly volume. (DataForSEO also supplies the data for the AI Answer Monitor and for Pro tracked-keyword ranking snapshots — see sections 4 and 5.)
• Cloudflare (privacy policy) — hosts the Worker, Workers KV cache, database and edge rate-limiting that power this tool. Cloudflare acts as a processor on our behalf.

Search Seasonality does not send your keywords to any AI or language-model provider. We never sell or share your keyword data with any party other than the processors listed above.

Sensitive keywords: because your keywords are transmitted to DataForSEO to produce the scan, please avoid using the tool to look up personal information, medical conditions, or any other information you would not feel comfortable sharing with that third party under its policy.

4. AI Answer Monitor (server-side data flow)

Like Search Seasonality, the AI Answer Monitor runs partly on our servers because it consumes data from a third-party SEO provider. This section explains exactly what is sent where. The tool has three modes with different data flows: the AI-answer exposure report, available on the free tier, and topic scans and prompt checks, which are Pro. All three require a signed-in account.

AI-answer exposure report (free tier)

What you provide:

• Your domain (e.g. example.com) — entered by you on the report form.
• A country — which Google market to measure.
• Brand terms (optional) — substrings used to exclude your own brand searches from the figures.

What is sent to the data provider: your domain and the country, in two requests to DataForSEO, which return the search terms your site ranks for and which of those carry an AI Overview. Brand terms are not sent to DataForSEO — the filter is applied on our servers to data already retrieved, which is why changing it costs nothing and returns instantly.

What we store:

• Report cache (up to 30 days) — the computed figures are stored in Cloudflare Workers KV under a key derived from the domain, the country and the number of terms requested, so repeat requests for the same domain are served without re-querying DataForSEO. The cached entry contains no user identifier, and it is shared: a report for a given domain may be served from a copy produced by another account's request.
• A daily counter — your user id and the current date, used to enforce the free daily allowance. It records how many reports you have run, not which domains.

The exposure report is not persisted to our database, and no report history is kept against your account.

Note on domains you do not own: the report accepts any domain, because it reads only public search-results data. If you enter a competitor's domain, that domain is sent to DataForSEO in the same way your own would be.

Topic scans and prompt checks (Pro)

What you provide:

• A search keyword (e.g. "best CRM") — entered by you on the scan form.
• An AI engine selection — ChatGPT or Google AI Overview.
• A country selection (Google AI Overview only — ChatGPT scans are US-English at the data provider's restriction).
• A search scope (optional) — Questions, Answers, or Both, controlling where within the AI engine's cached prompts the data provider looks for your keyword.

Free users cannot trigger a topic scan; they see a fixed sample (a hard-coded "best CRM" result). No data is transmitted to any third party in that sample path. This applies to the topic scan only — the free AI-answer exposure report described above does send your domain to DataForSEO.

What we process server-side (Pro scans only):

• Your authenticated user id — used to enforce your monthly scan quota (50 scans per month on Solo, 250 on Agency) and to attribute the persisted scan row to your account.
• Scan results cache (7 days) — once a scan completes, the full result (the keyword, country, platform, search scope, brand entities, fan-out queries, cited sources, AI search-volume enrichment, and DataForSEO cost telemetry) is stored in Cloudflare Workers KV for 7 days so that a second scan of the same tuple within the week is served from cache without re-querying DataForSEO. The cached payload does not include any data that identifies the user who triggered the original scan.
• Persisted scan history (D1 ai_mention_scans) — every completed Pro scan is also persisted to D1 alongside your user id, with the full DataForSEO payload in a raw_json column so the tool can re-render historical scans without re-paying DataForSEO. Failed scans, and scans where DataForSEO returned zero items, are not persisted (they also do not decrement your monthly quota).

Third parties that receive AI Answer Monitor scan data:

• DataForSEO (privacy policy) — receives your keyword, country code, AI engine, and search scope; queries DataForSEO's LLM Mentions endpoint, which returns brands, fan-out queries, and cited sources extracted from the AI engine's cached responses to prompts containing your keyword. Also queries the AI search-volume endpoint for the fan-out queries returned. DataForSEO does not receive your email or any other account identifier. (Note: DataForSEO does not query ChatGPT or Google's AI Overview live on your behalf — it serves data from its own cached index of prior third-party prompts.)
• Cloudflare (privacy policy) — hosts the Worker, Workers KV cache, D1 database, cron triggers, and edge rate-limiting that power this tool. Cloudflare acts as a processor on our behalf.

We never sell or share AI Answer Monitor scan data with any party other than the processors listed above. The brand names, source URLs, and fan-out queries returned by DataForSEO and stored in our database are not your personal data — they are public third-party data about the AI engine's response patterns — but they are scoped to your account so only you can see your own scan history.

Pro: watched scans and weekly diff cron. Weekly re-runs are paused at present (since 30 September 2026): nothing is re-scanned or sent on your behalf until they resume. When you click the ★ Watch button on a completed scan, we add the (keyword, country, platform) tuple to a "watch list" (D1 ai_monitor_watches, capped at 20 entries on Solo and 50 on Agency). Every Monday at 07:00 UTC, an automated cron re-runs each watched scan and computes a diff against the previous week's results (new and dropped brand mentions, new and dropped cited sources). The diff is persisted to D1 (ai_monitor_diffs) and surfaced on your /app/welcome dashboard. These cron-driven scans draw from a separate processing budget and do not decrement your manual monthly quota. The third-party data flow for each cron-driven scan is identical to a user-triggered scan (see "Third parties" above).

Pro: AI overview (optional). Click the "Generate AI overview" button on a completed scan and we send the scan's brand / fan-out / source tables to Anthropic to produce a 2-4 sentence plain-English summary. See section 5 for the shared AI overview data flow that covers Spiral, Timing Grid, AI Answer Monitor, and SERP Seasonality Map.

Sensitive keywords: because your keyword is transmitted to DataForSEO (and, if you click Generate AI overview, to Anthropic), please avoid using the tool to look up personal information, medical conditions, or any other information you would not feel comfortable sharing with those third parties under their respective policies.

4a. AI Answer Stability (server-side data flow)

AI Answer Stability runs on our servers, because it asks third-party AI engines the same question several times.

What you provide: a question (e.g. "What is the best server-side tagging platform?"), your brand name, and optionally some competitor names.

What is sent where:

• Your question is sent to DataForSEO (privacy policy), which puts it to ChatGPT (OpenAI), Gemini and AI Mode (Google), Perplexity, and Claude (Anthropic), several times each, and returns their answers and the sources they cited.
• Each answer, with your question, is sent to Anthropic (privacy policy) to list the brands the answer offers as options.
• Your brand and competitor names are not sent to anyone. They are matched against the answers on our servers.

What we store: the question, the brand names you entered, and the answers, cited sources and extracted brands, in our database (Cloudflare D1), against your account. If anyone asks the same question within 24 hours, they are shown the same answers rather than new ones; your brand names and your account are never shown to them.

Sensitive questions: because your question is sent to the providers above, please don't use the tool with personal information about anyone.

5. DATA HIT Tools Pro accounts

The Pro tier adds account-based features on top of the three free tools. Using Pro requires an email address (for sign-in) and a payment method (for the subscription). Everything in this section applies only to users who have created a Pro account; the free tools remain anonymous and unchanged.

What we collect when you create a Pro account:

• Email address — used as your account identifier and the destination for sign-in links and weekly digest emails. Stored in our Cloudflare D1 database (the users table). Never sold, never used for marketing without explicit opt-in.
• Stripe customer + subscription identifiers — opaque IDs returned by Stripe after checkout, stored alongside your email so we can match your account to your billing record on subsequent Stripe webhook events.
• Plan tier and status — pro_solo or pro_agency with one of trialing / active / past_due / canceled, kept in sync via Stripe webhooks.

We do not collect your name, address, or any other contact information for the Pro tier. Stripe handles all payment details directly — we never see your card number or full billing address.

Authentication (magic links):

• Sign-in is passwordless via a single-use email link. When you request a sign-in, we generate a random 256-bit token, hash it with SHA-256, and store only the hash in our database (table magic_link_tokens). The plaintext token is sent to your inbox via Resend (see "Third parties" below) and is valid for 15 minutes and one use only.
• After you click the link, we issue a session by setting an HMAC-signed cookie named dh-session (HttpOnly, Secure, SameSite=Lax, 30-day sliding TTL). The session row lives in D1 (table sessions); the cookie carries only the session id, not your email or plan tier.

What we store as you use Pro features:

• Tracked keywords (D1 keywords) — the keyword + country pairs you opt to track for weekly rankings (the Track button in Search Seasonality), scoped to your user id. Up to 20 active rows on Solo and 200 on Agency. Archived rows are soft-deleted, not erased, so historical snapshots stay attributable.
• AI Answer Monitor scans (D1 ai_mention_scans) — every Pro AI Answer Monitor scan you run is persisted with the keyword, country, AI engine, search scope, brand entities, fan-out queries, cited sources, and DataForSEO cost telemetry. See section 4 for the full data flow. Counts toward your monthly scan quota (50 Solo / 250 Agency, resets 1st of month UTC).
• Watched AI scans (D1 ai_monitor_watches) — (keyword, country, AI engine) tuples you have pinned via the ★ Watch button. Up to 20 on Solo, 50 on Agency. The Monday 07:00 UTC cron iterates this list to re-scan and diff (see "What we do with this data" below). Soft-deleted on unpin.
• Weekly AI mention diffs (D1 ai_monitor_diffs) — one row per (watched scan, week) recording the new brand mentions, dropped brand mentions, new cited sources, and dropped cited sources versus the previous week's scan. Surfaced on your /app/welcome dashboard. Hard-deleted with the parent watch on unpin or account closure.
• Weekly SERP snapshots (D1 snapshots) — every Monday at 06:00 UTC a cron job submits your tracked keywords to DataForSEO via their asynchronous postback API. The top-10 result list returned by DataForSEO is stored against the keyword id along with the capture timestamp. Snapshots are retained for the lifetime of the keyword and are deleted when the parent keyword is hard-deleted at account closure.
• Saved keyword sets (D1 saved_dashboards, with Search Seasonality digest records in seasonality_digest_events) — a set's name, country and keyword list, and which of its keywords the Monday email has already mentioned (keyword, country, month). Kept until you delete the set or close your account.
• Saved view configurations (D1 saved_dashboards) — when you click "Save current view" on a Pro preview, we persist the control settings (metrics, filters, date range, colour scale, etc.) as a JSON blob scoped to your user id and the tool. Up to 5 per tool on Solo, 50 per tool on Agency. Soft-deleted on archive.
• Annotations (D1 annotations) — short notes you log against a calendar date (or date range) that surface across all three tools. Label and optional body text are stored verbatim against your user id. Unlimited on both Solo and Agency. Soft-deleted on archive.
• Weekly digest history (D1 weekly_digest_sends) — one row per user per week recording whether the Monday-morning digest was sent, skipped, or failed, plus the Resend message id of the email. Used for idempotency (so a cron retry can't double-send) and for the "Most recent digest" line on your dashboard.
• Anomaly alerts (D1 anomalies) — when the Monday cron detects an analytics value that deviates significantly from its year-over-year baseline on the Seasonal Spiral, we record the date, metric, GA4 property id, and summary statistics (the actual value, expected value, sample standard deviation, and z-score). We do not store your raw analytics data; detection re-fetches from Google on every cron run. Rows are scoped to your user id and soft-deleted when you dismiss them. The scope of GA4 access used for detection is analytics.readonly — same as the persistent connection described in section 2.
• Billing event log (D1 billing_events) — every Stripe webhook we receive is recorded by provider event id for dedupe + audit. The payload is the raw Stripe event; we don't enrich it with anything we don't already have on your record.
• Search Console connection (D1 gsc_connections) — if you opt to enable the persistent Search Console connection on Spiral / Grid, or Search Console import in Search Seasonality, we store the OAuth refresh token issued by Google when you authorise our app. The refresh token lets us query Google's Search Console API on your behalf (e.g. for the Monday digest cron). We do not store the access token (it lives in memory for the duration of one API request). The scope granted is webmasters.readonly only.
• GA4 connection (D1 ga4_connections) — same shape as the Search Console connection above, but for GA4. Stored when you opt into the persistent connection on Spiral / Grid via "Stay connected next time". The scope granted is analytics.readonly only. Revocation works the same way: click "Disconnect" in the tool or visit your Google account permissions page; revocation soft-deletes the row and we no longer receive access.
• GA4 Tracking Health (D1 ga4_monitors, ga4_monitor_events, ga4_monitor_series, ga4_alerts) — only if you turn on tracking checks. We use your persistent GA4 connection to run automated, read-only checks once a day, at around 09:00 in your property's time zone (retried hourly if Google is unavailable). When sessions look like they have collapsed, we also make a real-time check of how many users are active right now. We store: the property you chose, with its name and time zone; the names of the events you choose to watch; daily totals for those events (counts, and how many were credited to the Direct, Unassigned and (not set) channels); daily session counts by platform (web, iOS, Android); and the alerts we raise, with the figures behind them and any feedback you give. We do not store page-level, user-level or raw event data. Daily totals are kept for 90 days on a rolling basis. Alert emails go to your account email address via Resend and always include a link to turn them off. Explanations in alerts are written from fixed templates: your analytics data is not sent to any AI provider for this feature, and we will update this policy before that changes. Turning tracking checks off, or switching to another property, deletes the stored totals and alerts for that property. The scope used is analytics.readonly.

What we do with this data:

• Render your Pro dashboard at /app/welcome with your tracked keywords, watched AI scans, weekly AI mention diffs, saved views, recent annotations, and most recent digest status.
• Submit your tracked keywords to DataForSEO weekly (Monday 06:00 UTC cron) so your dashboard and weekly digest stay up to date.
• Re-scan your watched AI Answer Monitor tuples weekly (Monday 07:00 UTC cron) and compute the brand / source diff versus the previous week. These cron-driven scans draw from a separate budget and do not decrement your manual monthly quota.
• Send you the weekly digest email each Monday morning at 08:00 UTC (skipped automatically when there is nothing to report — quiet weeks don't generate inbox noise).
• Enforce your plan-tier caps on inserts (tracked keywords, watched scans, saved views, AI scans per month) and authorise gated routes.
• Generate on-demand AI overviews of the visible chart on Spiral, Timing Grid and AI Answer Monitor when you click the "Generate AI overview" button (see the next paragraph for the data flow).

AI overview (on-demand Pro feature). The Spiral, Timing Grid and AI Answer Monitor offer a "Generate AI overview" button that produces a 2-4 sentence plain-English summary of the visible chart. The data we send to Anthropic differs per tool:

• Seasonal Spiral: the dataset label (e.g. "GA4 · 382578512"), the chosen metric (e.g. "sessions"), any active filter selections, and the array of (date, value) data points currently rendered. Capped at 730 points (~2 years of daily data).
• Timing Grid: the dataset label, metric, filter selections, and the 7×24 array of (day, hour, aggregate value) cells.
• AI Answer Monitor: the keyword, country, AI engine, and the top brand / fan-out / source rows the scan returned (capped at 20 each).

In each case the request goes to Anthropic via the Claude API; the response is sanitised (only <strong> tags allowed) and rendered inline. Generated overviews are cached in Cloudflare Workers KV for 24 hours, keyed by a hash of the full input payload, so re-clicking Generate on identical chart state returns the cached output without a fresh API call. We never send your email, user id, or any other account identifier to Anthropic. Per Anthropic's published policy, API requests are not used to train their models by default.

We do not use your Pro account data for advertising, profiling, AI model training, or any purpose unrelated to operating the tools.

Third parties that receive Pro account data:

• Stripe (privacy policy) — receives your email and payment details directly via the Stripe Checkout and billing portal flows. Stripe is the payment processor and is acting as a processor on our behalf for the subscription record.
• Resend (privacy policy) — receives your email address and the rendered content of any magic-link or weekly digest email we send. Acts as a processor on our behalf.
• DataForSEO (privacy policy) — receives the keywords + countries you have opted to track for weekly rankings, once per week (Monday 06:00 UTC), to look up the latest SERP top 10. Also receives the keywords + countries + AI engine of your watched AI Answer Monitor scans, once per week (Monday 07:00 UTC), to re-scan the LLM Mentions endpoint. DataForSEO does not receive your email or any other account identifier.
• Anthropic (privacy policy) — receives the visible chart data each time you click "Generate AI overview" on the Spiral, Timing Grid or AI Answer Monitor (see the "AI overview" paragraph above for per-tool specifics). Anthropic does not receive your email, user id, or any other account identifier. Per Anthropic's published policy, API requests are not used to train their models by default.
• Cloudflare (privacy policy) — hosts the Worker, D1 database, KV store, cron triggers, and edge rate-limiting. Cloudflare acts as a processor on our behalf.

We never sell or share Pro account data with any party other than the processors listed above. We never combine your Pro data with any other identifier.

Closing your Pro account: to permanently delete your Pro account and all associated data (tracked keywords, SERP snapshots, AI Answer Monitor scans, watched AI scans, weekly AI mention diffs, saved views, annotations, anomaly alerts, sessions, digest history, billing event log entries scoped to your record), email hello@datahit.co from the address registered on the account. We will cancel any active subscription via Stripe and hard-delete your D1 rows. You can also self-serve cancel the subscription at any time from your dashboard via the Stripe billing portal — that ends future billing but leaves your account data in place until you also email us to request deletion.

6. Analytics

We use Plausible Analytics, a privacy-focused analytics service, to collect anonymous usage statistics about our tools pages. Plausible does not use cookies, does not collect personal data, and is fully compliant with GDPR, CCPA, and PECR. No Google user data, scan data, keyword data, or Pro account data is shared with Plausible. The only data collected relates to page views and lightweight custom events (e.g. that a scan was triggered, that a Pro annotation was created — never the keyword text or annotation content itself).

7. Cookies and local storage

The free tools at tools.datahit.co set no tracking cookies. We use the following browser localStorage items, all of which are purely UI preferences with no identifying value and no third-party visibility:

• dh-theme — your light/dark theme preference.
• dh-spiral-annotation-hint-dismissed — set to 1 after you dismiss the one-time "right-click any cell to annotate" hint on the Seasonal Spiral tool (Pro feature). Suppresses the hint on subsequent visits.

Pro accounts use one HTTP cookie — dh-session, HMAC-signed, HttpOnly, Secure, SameSite=Lax, 30-day sliding expiry. The cookie carries only the session id; identity is resolved server-side. The cookie is not set unless you sign in. Signing out (from the dashboard or via the sidebar) deletes both the cookie and the server-side session row.

For information about cookies used on the main DATA HIT website, see our main privacy policy.

8. Data retention summary

Free tools:

• OAuth Google user data — never persisted; browser-memory only; discarded on disconnect, tab close, or token expiry.
• Search Seasonality volume cache — Workers KV, per keyword and country, until the next month's data is due (usually a few weeks), then automatically deleted.
• Share-link slugs (previous SERP Seasonality Map links) — stored in Workers KV for 30 days from creation, then automatically deleted.
• IP addresses — consumed transiently for rate limiting; never persisted.
• Browser preferences (theme, dismissed UI hints) — stored in your browser's localStorage until you clear it. See section 7 for the full list.

Pro accounts:

• Magic-link tokens — D1; SHA-256 hashed; 15-minute TTL; single-use. The hash is retained briefly past expiry for audit but never the plaintext.
• Sessions — D1; 30-day sliding expiry from last activity. Deleted immediately on sign-out.
• Tracked keywords — D1; retained until you archive them. Archived rows are soft-deleted (kept for snapshot attribution); hard-deleted on account closure.
• Weekly SERP snapshots — D1; retained for the lifetime of the parent keyword. Hard-deleted with the keyword on account closure.
• AI Answer Monitor scans — D1; retained for the lifetime of the account; the cached payload in Workers KV expires after 7 days. Hard-deleted on account closure.
• AI Answer Stability checks — D1; the question, your brand names and the answers, retained for the lifetime of the account and hard-deleted on account closure. Free accounts included.
• Watched AI scans + weekly diffs — D1; retained until you unpin the watch (which cascades the diffs) or close the account.
• AI overview text — cached in Workers KV for 24 hours from generation, then automatically deleted. The generated text is not retained in any other store.
• Saved view configurations and saved keyword sets — D1; retained until you delete them (soft-archive); hard-deleted on account closure, with the Search Seasonality email records.
• Search Seasonality set-scan log — D1; account id, time, country, keyword count and provider cost per keyword-set scan (not the keywords); hard-deleted on account closure.
• Annotations — D1; retained until you delete them (soft-archive); hard-deleted on account closure.
• GA4 Tracking Health — D1; daily totals for watched events and sessions are kept for 90 days on a rolling basis (including while checks are paused); alerts and your feedback are kept while tracking checks are on. Everything is deleted when you turn tracking checks off, switch property, or close the account.
• Weekly digest history — D1; retained for as long as the account exists for audit and dedupe.
• Billing event log — D1; retained for as long as the account exists for audit and reconciliation. After closure, your billing record at Stripe is retained per Stripe's own policy (typically 7 years for tax/regulatory reasons); we delete our local mirror.

To request earlier deletion of a specific cached scan, share slug, or entire Pro account, email hello@datahit.co.

9. Data protection

All communication between your browser, our Workers, and any third party we transmit data to is encrypted using HTTPS/TLS. Our Worker runs on Cloudflare's edge with strict secret-management for the API credentials used to call DataForSEO, Anthropic, Stripe, Resend, and (for OAuth tools) Google. The OAuth client IDs for our connected tools are restricted to authorised JavaScript origins, preventing use from unauthorised domains. Session cookies are signed with a server-side HMAC key so a tampered cookie cannot impersonate a Pro user; Stripe webhooks are verified with Stripe's HMAC-SHA256 signature scheme before any database write.

10. Your rights

Under UK GDPR and the Data Protection Act 2018, you have the right to:

• Access the personal data we hold about you
• Request correction of inaccurate data
• Request deletion of your data
• Object to or restrict the processing of your data
• Withdraw consent at any time

Free tools: Seasonal Spiral, Timing Grid and Search Seasonality (for a single keyword) require no account, and their cached data is short-lived and not tied to your identity, so most of these rights are fulfilled by design for those tools. The AI Answer Monitor is the exception: it requires a signed-in account even on the free tier, so the account rights described above apply to it in full. To revoke OAuth access, use the tool's disconnect button or your Google Account permissions page. To purge a cached SERP scan or share slug, email hello@datahit.co.

Pro accounts: you can review the bulk of what we hold about you directly from the dashboard at /app/welcome (tracked keywords, saved views, annotations, recent digest activity). To request a full export of your account data, to correct anything inaccurate, or to delete the account entirely, email hello@datahit.co from the address registered on the account.

If you have any concerns about how your data is handled, you have the right to lodge a complaint with the Information Commissioner's Office (ICO).

11. Changes to this policy

We may update this privacy policy from time to time. Any changes will be reflected by updating the "Last updated" date at the top of this page. We encourage you to review this policy periodically. Continued use of our tools after changes are posted constitutes your acceptance of the updated policy.

12. Contact

If you have any questions about this privacy policy or how your data is handled, please contact us:

• Email: hello@datahit.co
• Address: DATA HIT Ltd, Department, 4 The Boulevard, Leeds, LS10 1PZ

---