Privacy Policy — Kintsugi AI

Last updated: 7 October 2026 · Effective date: 7 October 2026

This Privacy Policy explains how Adrenalina Marketing S.r.l. ("we", "us", "our"), the developer of the Kintsugi AI – 404 to 301 Shopify application (the "App"), collects, uses, stores, and protects information when a merchant ("you") installs and uses the App.

By installing or using the App, you agree to the practices described in this policy. If you do not agree, please uninstall the App.

1. Who we are (Data Controller)

Adrenalina Marketing S.r.l.
Registered address: Corso Garibaldi 468/a, 89127 Reggio Calabria, Italy
VAT (Partita IVA): IT03226630808
Contact e-mail: amministrazione@ewebite.it

2. Scope of this policy

This policy applies only to the Kintsugi AI App and the data it processes through the Shopify Admin API. It does not apply to Shopify itself, to your storefront, or to any third-party service not listed below.

3. Information we collect

3.1 Store information

Your shop domain (e.g. your-store.myshopify.com).
The access token issued by Shopify when you install the App, used to call the Shopify Admin API on your behalf.
Your plan and subscription status for billing purposes.

3.2 Catalog and URL data

Product data: titles, handles, descriptions, vendors, product types, tags, prices, and URLs.
Broken (404) URLs detected on your store, including those found via your sitemap and internal links.
Redirect suggestions generated by the App and the 301 redirects created in your store.

3.3 404 tracker (optional app embed)

The App offers an optional theme extension, the Kintsugi 404 tracker. You enable it yourself in Online Store › Themes › Customize › App embeds; it is off until you turn it on, and turning it off stops all collection immediately. When enabled, a small script on your storefront's 404 page tells us that a visitor reached a missing URL, so that you can see which dead pages real people hit instead of only the ones a crawler can guess.

This is the only part of the App that involves your storefront visitors, so we state exactly what happens to each item:

Item	What we do with it
The missing URL path (e.g. /products/old-item)	Stored, together with a counter of how many times it was reached and the date it was first and last reached.
The referring page, if the browser sends one	Stored — one most recent sample per missing URL, so you can tell a broken internal link from an old external one. It is not linked to a person, a session, or any identifier, and we keep one value per URL rather than a history.
IP address and browser User-Agent	Used in memory, then discarded — never written to our database. They are read only to tell an automated crawler from a human visitor, because counting crawler traffic as customer demand would make the figures you act on meaningless (when we measured this, 96.6% of hits came from a single crawler). Our application does not log them.
Names, e-mail addresses, orders, carts, cookies, device identifiers	Not collected. The tracker sends no identifier of any kind, sets no cookie, and reads no storage on the visitor's device.

If you enable the tracker, you are collecting this through your own storefront. Depending on where your customers are, your own privacy notice and cookie/consent setup may need to mention it. We are telling you this plainly so you can decide: the feature is useful, and it is not worth a compliance surprise. If in doubt, leave it off — the App still works, using your sitemap, your deleted products and Search Console.

3.4 Google Search Console data (optional)

If you connect the App to your Search Console property, we read, for the URLs of your own site: the page URL, its impressions and clicks over a rolling window, and Google's indexing verdict for that URL (for example “Not found (404)”). This tells us which dead pages still receive visibility, so that work starts with the ones that cost you traffic.

Our access is read-only (Google scope webmasters.readonly): the App cannot change anything in your Search Console.
We request page-level data only. We do not request the search-terms dimension, so the queries your visitors typed are never sent to us.
You connect and disconnect this yourself; the App works without it.

3.5 Support data

The e-mail address and message content you submit through the in-app support form.

3.6 Preferences

Interface preferences (theme and language) stored locally in your browser via localStorage. These never leave your device and are not used for tracking.

We do not access your customers' personal data through Shopify. The App requests only product, navigation, and content permissions. It does not request access to customers, orders, carts, or any shopper personal information through the Shopify APIs, and therefore does not receive or store it.

The one exception, stated so this page does not contradict itself, is the optional 404 tracker in section 3.3: when you enable it, a visitor's IP address and User-Agent are processed in memory to separate humans from crawlers and are then discarded, and a sample referring page is stored per missing URL. Nothing there identifies a shopper to us.

4. How we use information

To detect 404 errors and match each broken URL to the most relevant 301 redirect target.
To generate AI-assisted semantic summaries and embeddings of your products to power matching.
To create and manage 301 redirects in your store when you approve them (or automatically, when you enable auto-apply).
To rank broken URLs by the traffic they still receive, using the 404 tracker (section 3.3) and Search Console (section 3.4) where you have enabled them.
To distinguish automated crawler traffic from human visitors, so that the volume figures you act on reflect customers rather than bots.
To monitor the App's own health and quality — counters, metrics, the URLs involved and excerpts of our operational error log are reviewed automatically each night, and a summary may be e-mailed to you. This never includes shopper data.
To provide customer support and respond to your requests.
To manage your subscription and billing through Shopify.

5. AI processing and sub-processors

To generate redirect suggestions, product text (such as titles, descriptions, and metadata) is sent to AI providers that create semantic summaries and embeddings. We share data only with the service providers below, strictly to operate the App:

Sub-processor	Purpose	Data shared
Shopify Inc.	App platform & billing	Store domain, catalog, redirects
Anthropic, PBC (Claude)	AI product summaries; automated review of the App's own health	Product text/metadata; counters, metrics, store URLs and excerpts of our operational error log
Voyage AI	Semantic embeddings	Product text/metadata
Resend	Transactional e-mail (support, account, reports)	E-mail address, message content
DigitalOcean	Cloud hosting & database	All App data, at rest

Google LLC is listed separately because the direction of the data is the opposite: we do not send Google your data, we read your own Search Console property on your behalf, read-only and only if you connect it (section 3.4). Your use of Search Console is governed by your own agreement with Google, not by this policy.

We do not sell your data, and we do not use it to train our own or third-party AI models.

6. Data retention

We retain your data for as long as the App remains installed. When you uninstall the App, or upon a verified deletion request, we delete your store record, catalog data, suggestions, and redirects from our database within 30 days. Support records may be retained longer where required for legal or accounting purposes.

6.1 Backups

We keep nightly backups of the App's database so that a failure cannot destroy your data. On our server they are kept on a rolling basis — the 7 most recent, after which the oldest is deleted. We also hold one copy outside the server, against the failure of the server itself; that copy is replaced by the newer one rather than accumulated, is transferred over an encrypted connection and is stored with restricted access. A deletion request is applied to the live database immediately and works through the backups as the rolling window passes, within the same 30-day period.

7. Data security

All data is transmitted over encrypted HTTPS connections.
Access tokens and credentials are stored with restricted access; admin passwords are hashed (PBKDF2-HMAC-SHA256) and never stored in plain text.
Access to production systems is limited to authorized personnel.

8. Your rights (GDPR and similar laws)

If you are located in the European Economic Area, the United Kingdom, or a jurisdiction with similar laws, you have the right to access, rectify, erase, restrict, or port your personal data, and to object to processing. To exercise these rights, contact us at amministrazione@ewebite.it.

We support Shopify's mandatory compliance webhooks (customers/data_request, customers/redact, shop/redact). Because the App does not store shopper personal data, data-request and customer-redaction webhooks return no personal data; the shop-redaction webhook triggers deletion of your store's data from our systems.

9. International data transfers

Some of our sub-processors are located outside the European Economic Area (for example, in the United States). Where data is transferred internationally, we rely on appropriate safeguards such as the data processing terms and standard contractual clauses offered by those providers.

10. Cookies and tracking

The App does not use advertising or analytics cookies, and does not track you across sites. The optional 404 tracker described in section 3.3 sets no cookie and reads no storage on a visitor's device. Inside the app, your own browser's localStorage holds two kinds of thing:

Your preferences — theme, language, owner/specialist mode, the reporting period you last chose, and which one-off notices you have dismissed.
A short-lived cache of your own store's data, so the app does not re-fetch the same list every time you switch tab: the current page of broken URLs and their proposed targets, your counters, and your latest weekly summary. It expires after 5 minutes, it is your own data shown back to you, and it never leaves your browser — we cannot read it, and it is not sent to any third party. Clearing your browser data removes it; the app then simply asks the server again.

The administration panel additionally uses short-lived session tokens.

10.1 Server logs

Like any web service, the web server in front of the App writes standard logs, which contain the IP address of the requesting client. They exist to diagnose faults and identify abuse, are never used to profile anyone, are not combined with the App's data, and are not shared.

The access log is rotated every day and we keep 14 previous days, so an entry is deleted within about 15 days.
The error log follows the same rotation, but a day on which nothing went wrong produces no file to rotate. It therefore holds its 14 generations over a longer span — typically a few weeks — because errors are rare. It is the smaller of the two by orders of magnitude, and each entry is deleted once 14 further rotations have occurred.

11. Children

The App is a business tool intended for merchants and is not directed to children under 16.

12. Changes to this policy

We may update this Privacy Policy from time to time. Material changes will be reflected by updating the "Last updated" date above. Continued use of the App after changes take effect constitutes acceptance of the revised policy.

13. Contact us

For any question about this policy or your data, contact:
Adrenalina Marketing S.r.l. — amministrazione@ewebite.it