Privacy policy

Summary

The short version: the answerproof.ai website collects very little — an email address if you sign up for updates, standard server logs, and Google Analytics usage data. The Shopify app reads your store's public content and catalog to grade it, writes to your products only when you approve a specific change in the app, keeps your recent scan results and a before/after record of every change you apply, and deletes all of it when you uninstall.

Payment for paid plans is handled by Shopify; we never see card or bank details. Nothing is sold or shared for advertising, and you can ask us to correct or delete your data at any time — see “Your rights” below.

Who we are and how to contact us

AnswerProof is operated by Joe Caridi LLC, the Shopify Partner behind the AnswerProof apps. For any question about this policy or about your data, email [email protected].

What this policy covers

This policy covers three things: the AnswerProof AI Visibility app for Shopify (the "app"), on every plan — Free, Growth, and Pro; the AnswerProof AI readiness check browser extension (the "extension"); and the public answerproof.ai website (the "website"). Your relationship with Shopify itself — your store, your Shopify account, your customers' data, and your subscription payments — is governed by Shopify's own privacy policy, not this one.

What the app does

You install the app from the Shopify App Store and approve its access through Shopify OAuth. Depending on your plan, it does up to five things:

• Audit (every plan, including Free). Reads your store through the two read-only paths described below and produces a scored report. A scan never writes anything to your store.
• Product Review (Growth and Pro; a limited single-product version on Free). Scores each product's content for AI readiness and can generate AI-written description rewrites, structured “spec sheet” data, and schema.org JSON-LD for products. Nothing is written to your store unless you review and explicitly approve that specific change in the app.
• Hosted llms.txt (Growth and Pro). Generates a plain-text summary of your catalog and serves it on your own domain at /apps/answerproof/llms.txt through a Shopify app proxy. It contains only public catalog content.
• AEO Performance, or citation monitoring (Pro). You pick up to 50 of your own pages; once per billing period the app asks third-party AI engines a small set of buyer-style questions about those pages and stores each engine's answer so you can see whether your store was cited.
• Scheduled re-audits and alerts (Pro). The app re-scans your store weekly and emails you if your score drops by more than the threshold you choose.

Data we collect and store (the app)

The audit reads your store in two ways:

• Your public storefront. The app fetches roughly 20 public pages the way an AI crawler would — robots.txt, sitemap.xml, your homepage, up to 8 product pages, 2 collection pages, 2 blog articles, and Shopify's agentic endpoints such as agents.md and llms.txt. It identifies itself as AnswerProofAuditBot and fetches politely, with rate limits.
• The Shopify Admin API, using the read scopes read_products and read_content. Through them it reads product titles, descriptions, SEO fields, image alt text, product type, vendor and category, collection information, and blog articles, up to your 500 most recently updated active products.

For each shop, we store:

• your .myshopify.com shop domain (and your store's primary domain), install and uninstall timestamps, and the API access token Shopify issues so the app can operate — held in the app's session store together with the name and email address of the Shopify user who installed the app, as Shopify provides them;
• your scan results — scores, per-check findings with their evidence (short excerpts drawn from your store's public content and catalog), the generated report summary, and scan statistics;
• your products' AI-readiness scores and their history;
• draft rewrites awaiting your review, the spec notes you type, and the structured spec rows derived from them;
• a before/after record of every change you apply, kept so every change can be rolled back;
• your store's generated llms.txt;
• on the Pro plan: your monitored pages, the buyer questions the monitor asks, a snapshot of each AI engine's answer (kept so any citation result can be audited), and your re-audit alert settings, including the alert email address if you enter one — otherwise alerts go to the installing user's email address on file with Shopify;
• billing state: which plan is active, when your billing cycle turns over, your monthly usage counters (credits used), and a record of any credit top-up purchases. Payment itself is handled by Shopify — see “Billing” below.

We keep your 25 most recent scans.

Data we never collect

We never request, see, or store customer (shopper) data of any kind — no names, no email addresses, no physical addresses, no orders, no payment details. The app has no Shopify permissions that would expose customer data to it.

Shopify sends every app its customers/data_request and customers/redact privacy webhooks. We acknowledge them, and because we hold no customer data, there is never anything to return or delete under them.

How we use your data

We use the data above for exactly one purpose: producing your own reports, product scores, draft rewrites, citation results, alerts, and history, and applying the changes you approve. We do not sell your data, we do not use it for advertising, and we do not share one merchant's data with another.

Shopify permissions, and when the app writes to your store

The app requests exactly four Shopify permissions:

• read_products and read_content — the audit's reads, described above.
• write_products — used exclusively to apply changes you have explicitly approved in the app: product descriptions and answerproof.* product metafields (the spec sheet and JSON-LD). A scan never writes.
• write_app_proxy — used only to serve your hosted llms.txt on your own domain. Read-only content; no customer interaction.

Every write follows the same rule: you see the proposed change, you approve that specific change, and only then is it applied. Before every approved write the app snapshots the previous value so the change can be rolled back with one click. The app never edits anything you did not approve.

AI processing

Three things in the app are produced with Anthropic's Claude API: your report's executive summary, product AI-readiness scores, and description rewrites. What we send to Anthropic: the scan's findings (check results, scores, and evidence excerpts from your store's public content), your shop name, and — for Product Review — your own product content: titles, descriptions, options, and the metafields you typed. What we never send: customer data, your storefront password, or your API access token. Summaries are cached so identical findings are not sent again.

Every draft rewrite is checked by a deterministic grounding step against the product's own data, and no rewrite is published without your explicit approval in the app.

On the Pro plan, if you use citation monitoring, the monitor sends each monitored page's buyer-style questions to third-party AI engines — OpenAI, Perplexity, and Anthropic — and stores their answers. Those questions are about your pages; they never include customer data or your storefront password. If you do not use citation monitoring, nothing is sent to OpenAI or Perplexity.

Subprocessors and infrastructure

We rely on a small set of providers to run AnswerProof:

• Shopify — the platform through which all store data is accessed and through which subscriptions are billed.
• Anthropic — generates the report's executive summary, product AI-readiness scores, and description rewrites, and serves as one of the citation-check engines on Pro.
• OpenAI and Perplexity — citation-check engines, used only for Pro merchants who use citation monitoring, as described above.
• SMTP2GO — delivers the score-drop alert emails on Pro, sent from an answerproof.ai address to your alert address. Those emails contain only the recipient address, your store name, and the score figures.
• Railway — hosts the app and its PostgreSQL database.
• Cloudflare — serves the answerproof.ai website and routes our email.
• Google — provides the Google Analytics service that measures traffic on the answerproof.ai website.

Billing

The Free plan costs nothing. Growth and Pro are paid subscriptions, and credit top-ups are one-time purchases; both are billed by Shopify through its app billing system. Shopify handles the payment itself, and the app never sees your card or bank details. The app keeps only the billing state it needs to operate: which plan is active, when your billing cycle turns over, your monthly usage counters (credits used), and a record of any credit top-up purchases.

Data retention and deletion

We keep your 25 most recent scans per shop. If you uninstall the app, Shopify notifies us through its standard privacy webhooks, and your stored data is deleted as part of Shopify's standard redaction process: when the shop/redact webhook arrives (about 48 hours after uninstall), your shop record and everything attached to it — scans, summaries, product scores, draft rewrites, spec notes and spec rows, the applied-change history, your generated llms.txt, monitoring data, alert settings, and usage counters — is permanently deleted. Content you approved and applied to your own products stays on your store, because it is your product content.

You can also request deletion at any time by emailing [email protected].

Storefront password handling

If your store is behind a storefront password (for example, before launch), you can choose to enter it so the app can scan your storefront. Your password is used once, in memory, for that single scan: it is never written to our database, never logged, never included in your scan results, and never sent to any third party. Each new scan asks for it again.

The browser extension

The AnswerProof AI readiness check extension collects nothing and sends nothing. Every check runs entirely on your device.

When you click the extension's icon, it reads the page in your current tab — headings, paragraphs, metadata, and structured data — and computes its report locally, in your browser. In detail:

• It makes no network requests to us or to anyone else. The one outbound request it triggers is for the current site's own public robots.txt file, made by your browser directly to that site, to report whether AI crawlers are allowed there.
• It stores nothing — no browsing history, no page content, no settings. Nothing persists after the popup closes.
• It has no account, no cookies, and no analytics or tracking of any kind.
• It only runs when you click its icon. It never reads pages in the background.

The extension's link to this website carries campaign tags (utm_ parameters) that tell us the visit came from the extension; they identify the extension, not you. Once you are on answerproof.ai, the website terms below apply.

Website visitors

The public answerproof.ai website uses Google Analytics to understand which pages people visit and how they found us. Fonts are served from our own server, not a third party.

• Google Analytics sets cookies and collects usage data: pages viewed, approximate location derived from your IP address, browser and device details, and the site that referred you. Google LLC processes this data on our behalf. You can block these cookies in your browser settings or with Google's opt-out browser add-on.
• If you enter your email address in the signup form on this site, we store the address you enter, the time you signed up, your browser's user-agent string, and any campaign tags (utm_ parameters) in the link that brought you here. We use this only to send you the AnswerProof updates you signed up for.
• To limit abuse of that form, we count recent signup attempts per IP address in server memory; these counts are never written to disk.
• Like most websites, our server keeps standard technical logs (IP address, requested page, time) used only for security and troubleshooting.
• The website is served through Cloudflare, which processes visitor traffic (including IP addresses) as our network and security provider.
• If you create an account and sign in to the AnswerProof Web product, the site sets a session cookie strictly necessary to keep you signed in. There are no advertising cookies anywhere on the site.

Your rights

You can ask us at any time to access, correct, or delete the data we hold about you or your shop by emailing [email protected]. Depending on where you live, laws such as the EU/UK GDPR or the California CCPA may give you formal versions of these rights, including the right to complain to a data protection authority. We honor access, correction, and deletion requests from everyone, regardless of location.

Changes to this policy

If we make material changes to this policy, we will update this page and the “Last updated” date at the top. The current version always lives at this address. This policy took effect on July 15, 2026. It was last updated on September 11, 2026, to describe the paid Growth and Pro plans — merchant-approved product writes, the hosted llms.txt, citation monitoring, and scheduled re-audit alerts — and the additional processors they use: OpenAI, Perplexity, and SMTP2GO.

Contact

Email [email protected]. Mail to any answerproof.ai address is forwarded (via Cloudflare Email Routing) to the operator's inbox and is used only to answer your inquiry.