The answers you submit
The free matching flow processes the style and shopping answers you submit: design direction, shape, metal, wear, setting priorities, diamond preference, budget, spending priority, features to avoid, and an optional note.
Those answers are sent to the RingMatcher server only when you request a Ringprint. They are used to create the result and check that the recommendation respects the choices you made.
Optional image handling
Photos are optional and face photos are never required. Selected images are previewed and resized in your browser. Up to three resized images and their labels may be sent through the RingMatcher server to its configured AI provider for that request. If image analysis is unavailable, your answers alone create the result.
RingMatcher does not intentionally save uploaded image files to an account or image database. Zero-data-retention and data-collection-denial settings are requested from the configured provider, but that provider’s infrastructure, terms, and legal obligations still apply. Do not upload an image you are not comfortable transmitting for analysis.
Sensitive inferences are outside the purpose
The image-analysis instructions restrict the model to visual design clues such as color, line, proportion, texture, silhouette, accessories, objects, and interiors. They prohibit identifying a person, brand, location, or private detail and prohibit inferring sensitive traits.
Text or instructions visible inside an uploaded image are ignored. A user-supplied ‘not their style’ label is treated as negative evidence; a general style clue is treated as lower-confidence context rather than proof.
Shared links
A Ringprint share link contains the visible recommendation, selected diamond preference, and budget bracket in the URL fragment. It does not contain uploaded image files or the questionnaire answers as separate fields. Text you submit may influence the visible explanation, so review the result before sharing it.
Anyone with the link can read the information encoded in it. The public service does not create an account or a private, access-controlled workspace for the free Ringprint.
When you continue from a Ringprint to the retailer finder, the visible shape, setting, metal, center-stone range, origin, and budget bracket are transferred in a bounded URL fragment. A fragment is not sent with the retailer-page HTTP request or written to server access logs, and RingMatcher removes it from the address bar after importing it. Anyone who receives the unimported link can still read those fields.
Paid packs and Stripe
Optional paid packs use Stripe for one-time payment processing. When you start checkout, RingMatcher sends Stripe the receipt email you enter, the selected pack identifier, the server-set amount and currency, and generated request and PaymentIntent identifiers. Stripe’s Payment Element collects and processes the payment details. RingMatcher does not receive or store your full card number, security code, or the contents of Stripe-hosted payment fields.
Stripe metadata contains only generated identifiers, payment and fulfillment status, the pack identifier, and a one-way hash used to bind access to the browser that created the checkout. RingMatcher does not put questionnaire answers, uploaded photos, private notes, or paid report contents in Stripe metadata.
RingMatcher places the PaymentIntent identifier and a generated access capability in per-pack HttpOnly, SameSite cookies with a 100-day maximum age so the server can verify that browser’s payment directly with Stripe. These cookies do not create an account or portable, cross-device access. A pending-checkout record may also remain in that browser’s local storage for up to 30 days so a confirmed payment can be recovered after a refresh without beginning another charge. It contains the pack and PaymentIntent identifiers, receipt email, request identifier, status, timestamp, and whether the checkout replaces an older pack purchase—not card details, questionnaire answers, photos, or report contents.
Stripe’s infrastructure, terms, privacy policy, and legal obligations apply to its processing. Paid report sessions are currently held in the RingMatcher server process, so a restart or deployment may interrupt server-backed follow-ups or downloads even though Stripe still records the payment.
Request limits, first-party measurement, and technical logs
To prevent abuse, the server may use your network address to derive a temporary request-limit key held in server memory. The raw address is not included in your Ringprint, its share link, or an analytics event.
RingMatcher sends small, first-party events for funnel milestones across the landing page, Ringprint creation, sharing and downloads, optional checkout and paid-pack workflows, and the retailer finder. The event contract permits only the event name plus fixed source, channel, pack, status, or purchase-type values, booleans, bounded counts or scores, and reviewed catalog retailer IDs. It excludes questionnaire answers, budgets, destinations, notes, generated direction text, URLs, page paths, payment identifiers, error digests, messages or stacks, and arbitrary text.
Accepted events are written only as structured server deployment logs. RingMatcher does not send them to a third-party analytics service or store them in a dedicated analytics database; retention depends on the deployment’s logging configuration. Events do not include an account or user identifier. Technical AI usage such as model name, token counts, response status, and estimated processing cost may also be logged without questionnaire answers or uploaded photos.
What RingMatcher does not sell
RingMatcher does not sell rings or personal data. If accounts, payment practices or providers, analytics providers, storage, or model providers change, this page should be reviewed before the changed service is offered publicly.
For privacy questions or requests, email hello@ringmatcher.com.