SoulMatch Privacy Policy
Version date 9 October 2026
This version takes effect when published by SoulMatch.
This Policy explains what personal information SoulMatch collects, how we use and share it, how long we keep it and the choices and rights available to you. SoulMatch is a non-clinical emotional wellness and self-reflection service for adults; it does not provide therapy, diagnosis, treatment, crisis monitoring or emergency response.
We protect Private Content, including chats, reports, journals and their log copies. We do not sell it or use it for targeted advertising or public promotional stories. Internal training requires your separate affirmative authorization; saving history is not that authorization. Chat deletion does not automatically delete every separate record. Sections 4–6 explain these distinctions.
1 About this Policy
1.1 Responsible company
Soul Alchemy World Limited, a company incorporated in Hong Kong, is responsible for the processing described in this Policy. Soul Alchemy World Limited provides the SoulMatch AI service and operates the online stores at www.soulalchemyworld.com and www.soulmatch.store; the same entity stands behind both brands. Our internal team handles privacy enquiries and requests. Contact us at support@soulalchemyworld.com or Flat 2401-16, 24/F, Wing Shing Industrial Building, 26 Ng Fong Street, San Po Kong, Kowloon, Hong Kong. This Policy should be read together with our Terms and Conditions, which govern both the SoulMatch AI features and the soulalchemyworld.com store.
1.2 Scope
This Policy covers SoulMatch and related account, product, payment, subscription, ST-credit and partner features on www.soulalchemyworld.com and www.soulmatch.store. The current release includes AI conversations, emotional-pattern scans, generated reports, support tools and journals, wearable, space and visual companionship experiences, customized bracelets, history, shopping, ST credits and partner features where available. Human companionship, professional appointments and sending reports to a professional for a booking are not offered in this release.
1.3 Adults only
SoulMatch is intended only for adults aged 18 and older, consistent with the age requirement and age attestation in our Terms and Conditions (clause 2.1). We do not offer this release to minors and do not knowingly collect personal information from anyone under 18. During the first-use process after sign-in, we require users to attest that they are 18 or older before proceeding with the service. If we learn that a minor has provided personal information, we will investigate, restrict access as appropriate, and delete the information unless we are required to retain it, in accordance with applicable law; a parent or guardian may contact support@soulalchemyworld.com to request deletion. This release does not offer a dedicated minor mode.
1.4 Which practices apply to you
This Policy covers two kinds of data practices, and which ones apply depends on how you use our websites. Our store and commerce data practices — covering orders, shipping, delivery, account records, payments, store cookies and related transaction information — apply to users of both www.soulalchemyworld.com and www.soulmatch.store. Our SoulMatch AI-service data practices — covering Private Content, processing of your inputs by the Google Gemini API to generate AI outputs, internal AI training, and AI-derived consumer-health and emotional data — apply only to users of the SoulMatch AI service, which is available at www.soulmatch.store. www.soulalchemyworld.com is a store and does not host the SoulMatch AI service; it may link you out to that service, and if you follow that link and use it, the AI-service data practices in this Policy then apply to your use of the service. The protections for Private Content and sensitive personal information apply wherever we receive such information, including through either website, customer support or a purchase flow, even if you do not use the SoulMatch AI service. Regardless of which practices apply, your non-waivable privacy and consumer rights continue to apply.
2 Information we collect
2.1 Account transaction product and consent records
Depending on the feature you use, we handle account identifiers, email, authentication information and optional profile details; orders, payments, subscriptions, ST-credit transactions, customer-support submissions and partner-account records; and purchaser or recipient names, contact details and delivery addresses needed for purchases. We record your confirmation that you are at least 18, acknowledgement of service information, consent to sensitive-content processing where requested, and separate internal AI training choice and its authorization record. Saving a conversation is not training authorization. Customized bracelet orders include the selected design, wrist measurement, bead materials, sizes and arrangement, Classic or NFC mode, NFC bead color where applicable, optional gift message, design images and order information. We collect the email address you provide for pre-dispatch photo confirmation, your acceptance of the purchase notice with its version, language and time, and subsequent order correspondence. Purchase acknowledgement and photo confirmation are not marketing or training consent. Payment providers process payment details for the transaction.
2.2 Private Content
"Private Content" means chat messages, AI replies, emotional-pattern scan and reflection answers, mood selections, journal entries, generated reports, summaries, emotional patterns and other inferences derived from these materials, relevant sensitive wellness inputs, and feedback containing any of that content. It includes copies in conversation history, reports, support submissions and operational or safety logs, regardless of storage location or label. Content about you or another person may be sensitive even without a real name.
2.3 Operational conversation records
Backend records may include user messages and AI replies, conversation and message identifiers, language, selected mood, entry route and associated technical or safety information. Certain records include risk classifications and routing information. These records are separate from the history shown in your browser.
2.4 Technical information statistics and sources
We process information needed for requests, sessions, troubleshooting, usage measurement and safety, including IP addresses where recorded, device and connection information, request and session identifiers, URLs and entry routes, and relevant security records. First-party usage statistics are off by default and start only after user opt-in. When enabled, they use a random current-session identifier, path-only page views and pre-defined non-sensitive event names. We do not collect query parameters, page titles, raw button or link text, element IDs, arbitrary metadata, mood text, consultation content or other free text for first-party statistics.
Information comes from what you submit, your device and use of our services, AI-generated outputs and inferences, and transaction or service information returned by our payment, commerce and other relevant providers. A restriction on optional usage-statistics fields does not mean that operational conversation records contain no message text.
2.5 Speech input
Where browser speech recognition is available, it converts speech to text for your input through browser-provided functionality. SoulMatch does not deploy an independent voice-recognition service and does not receive or store raw audio recordings. Sent text is processed as conversation content. Browser speech services may process audio under their own arrangements.
2.6 Product design NFC and generated images
Wearable, space and visual experiences can process selected mood, intention, context or product preferences to generate or recommend combinations. Bracelet personalization can use the full birthday you enter, birth month, derived zodiac or birthstone recommendations and your selected intention or mood. Saved designs may include this context, selected materials, bead layout, wrist size, product mode, prompts and generated images. Design history is stored in the browser and can be synchronized to your account; pending synchronization records may also remain in browser storage. These are generated design images, not a feature for uploading photographs of your home or reference photographs. Standard NFC products open www.soulmatch.store; the current implementation has no tag-ID tracking, location feature or tag-to-account binding. Opening the site can still generate ordinary website or session records, and signing in or ordering uses the relevant account and transaction flow. When you tap an NFC-enabled product it opens www.soulmatch.store and may load account or personalized content, and these interactions are treated as service-interaction data under this Policy.
3 Why we use information
3.1 Providing the services
We process information to generate AI replies, emotional-pattern reflections, reports and personalized product or design outputs; maintain conversation context; save selected content and history; operate support tools; handle purchases, subscriptions, ST credits and partner features; and respond to support requests. For customized bracelets, this includes producing the agreed design, processing payment, communicating actual bracelet photographs before dispatch, recording your reply, arranging delivery and handling corrections, returns or other after-sales requests.
3.2 Operations safety and legal obligations
We process necessary records to troubleshoot failures, detect misuse, investigate incidents, enforce service rules, maintain platform integrity, process deletion requests, respond to lawful requests, prevent fraud and meet financial, accounting, tax or legal obligations. SoulMatch does not provide continuous human monitoring, emergency dispatch or proactive emergency-contact or guardian notification.
3.3 Purpose limitation
We use personal information only for the purposes described in this Policy, for purposes compatible with those purposes where permitted by applicable law, or for another purpose where we obtain the required notice, consent or authorization. Private Content is not used for targeted advertising, public testimonials, promotional stories or unrelated third-party marketing.
4 AI processing and training
4.1 AI generation
To generate a reply, scan interpretation, report or other AI output, we send the input and context relevant to that feature to its AI processor. AI conversation and report generation use Google Gemini Developer API. We do not use Doubao or the Volcano Engine Ark API for these functions. Gemini may use project-isolated implicit in-memory caching with a 24-hour expiry. This provider cache is separate from SoulMatch chat-history storage; it does not mean your SoulMatch conversations automatically expire after 24 hours.
4.2 Provider retention and safety processing
Google retains prompts, context and outputs for 55 days for abuse monitoring and required legal or regulatory disclosures. Authorized Google personnel may review flagged content. Google states that abuse-monitoring data is not used to train or fine-tune models other than models specifically for policy enforcement. These limited safety uses are separate from SoulMatch internal training and from general provider product improvement.
4.3 Internal training and your choice
SoulMatch uses Private Content for internal training, fine-tuning, evaluation, testing, human annotation and related AI improvement only with your separate affirmative authorization covering that content and purpose. The training option is off unless you actively select it or we restore a previously recorded affirmative choice. A legacy default-selected flag is not sufficient evidence of authorization. We record your choice and distinguish authorized and unauthorized conversation records in our internal management system. Refusing training authorization does not prevent ordinary AI conversation. Saving a conversation is not training authorization. Historical records without valid authorization are not eligible for training.
You may withdraw your authorization by deselecting the training option or contacting support@soulalchemyworld.com. Withdrawal stops future use within the scope of the withdrawal. For identifiable copies already included in training or evaluation materials, we will assess and carry out applicable deletion or exclusion requirements and explain any lawful retention limits. Deleting source material does not necessarily remove its effects from a model that has already been trained; we do not promise technical removal that we cannot perform.
Internal development has included real user conversations, reports and journals. A new acceptance or authorization does not retrospectively authorize earlier processing. The authorization must cover the relevant content and purpose, including reports or journals when used; a conversation setting alone does not automatically authorize every separately stored record.
4.4 Restrictions on provider use
We do not grant AI providers permission to use Private Content for unrelated marketing or their own general model improvement. Applicable paid Gemini API terms restrict Google from using prompts and responses to improve its products. The limited caching, abuse monitoring, legal disclosures and policy-enforcement uses described in Sections 4.1 and 4.2 still apply. Your SoulMatch training choice governs our internal development; it does not disable the processing necessary for an AI provider to deliver and secure a response. Provider terms are available at https://ai.google.dev/gemini-api/terms, with retention and safety details at https://ai.google.dev/gemini-api/docs/zdr and https://ai.google.dev/gemini-api/docs/usage-policies.
5 History and saving choices
5.1 Saving and continuing a conversation
Server saving for continuation is on by default unless you previously chose otherwise. You can turn it off in Personal Settings under chat history. Turning it off stops subsequent continuation saves; it does not delete existing records or change your separate training choice. There is no automatic 24-hour expiry for SoulMatch chat history. Retention remains subject to Section 6.
The history you can view or restore depends on the browser and session information available to the application. We do not promise automatic recovery of all conversations on a different device. Sharing an account or device may allow others to see available records.
5.2 Reports journals and memory
Reports are stored separately from chat history for review or export. Local report history, access flags and local journals may remain on your device until deleted or browser data is cleared. Deleting a conversation does not automatically delete these records or the separate records described in Section 6.2. A saved report is not a separate long-term preference profile or vector-memory feature; we do not currently offer an independent preference-memory editor.
6 Deletion and retention
6.1 Self service controls
You can view available browser conversation records, export a selected conversation as text, or delete one or all listed conversations. When the application has the session information for a corresponding server-continuation record, the deletion operation also requests removal of that record. A failed server deletion is shown for retry rather than being treated as completed.
6.2 Scope of deletion and broader requests
Deleting a conversation from chat history does not automatically delete saved reports, local journals, your account, saved bracelet designs, orders, ST or subscription records, separate operational and safety logs, provider abuse-monitoring records, or any authorized training copies. Deleting a saved design is also separate from deleting a paid order, payment-provider records or order correspondence. Other browsers, anonymous sessions and ordinary AI logs that are not linked to your account may not be located by an account-based deletion request.
For a broader deletion request covering other personal information, contact support@soulalchemyworld.com. We use reasonably available identifiers or session details to locate relevant records, explain any limitations and available alternatives, and handle requests within applicable legal periods. We do not require you to resend a complete sensitive conversation merely to verify a request. Where applicable, we coordinate deletion with recipients and processors; lawful exceptions are explained rather than treated as permission to retain every conversation.
6.3 Retention criteria and account closure
We retain personal information only for as long as reasonably necessary for the service, transaction, security or legal purpose for which it was collected. Saved history and reports may remain while your account is active and you keep the content, subject to deletion requests and applicable law. Account closure has a 30-day waiting period, after which account-data cleanup starts through the cleanup process. The waiting period is not a promise that all copies are deleted on day 30 and does not suspend statutory rights-request deadlines.
For operational and safety logs, we consider whether an incident or troubleshooting task remains open, the minimum evidence needed, applicable limitation periods and legal preservation duties. Consent records are kept as needed to demonstrate and enforce the relevant choice. Authorized training copies are retained only for the authorized development purpose, subject to withdrawal and applicable deletion requirements. Data no longer necessary for these purposes is deleted or made non-identifying; merely renaming it a log does not justify indefinite retention.
6.4 Specific retention periods
First-party usage-statistics logs are kept for 90 days by default and can be configured from 1 to 365 days. Gemini abuse-monitoring records may be retained by Google for 55 days. Transaction, payment, subscription, refund, chargeback, tax, accounting and related financial records are kept for the period required or reasonably necessary for legal, tax, accounting, audit, fraud-prevention or dispute purposes; where Hong Kong tax-record rules apply, necessary business records are retained for at least 7 years.
Order photographs, confirmation replies, gift messages and design details are retained for the relevant fulfillment, after-sales or dispute purpose under Section 6.3. The financial-record period does not mean every chat, full birthday or email attachment must be retained for seven years.
6.5 Backups and non identifying information
We do not currently maintain separately configured business-database backups, cloud-disk snapshots, cross-region backups or disaster-recovery copies. This does not mean that external processors keep no records; provider records are subject to their applicable arrangements and this Policy. If our backup arrangements change, we will explain the applicable retention and deletion rules.
We may retain aggregated or other information that cannot reasonably identify you. Nicknames, hashes, session identifiers and other linkable pseudonymous data are not treated as fully anonymous if they can still reasonably be associated with you.
6.6 Safety records
Operational or safety records can include risk-related categories, actions, timestamps, identifiers, hashes and relevant context. These records are not clinical diagnoses or promises of continuous monitoring. They are used only within a documented and necessary safety, security, legal, audit or service-integrity purpose, and are not used for targeted advertising or public promotional stories. Any separate use for internal AI development remains subject to the affirmative training authorization described in Section 4.3.
7 Recipients and human access
7.1 Service providers generally
Providers process only the information needed for their function: AI generation, hosting and storage, payment, commerce and delivery, email, customer support, or optional advertising and usage measurement. Listing providers together does not mean every provider receives every field. Processors are restricted by role, instructions, contract and applicable law.
7.2 Provider categories
Our main providers include Google Gemini API for AI processing; UCloud for application hosting in Los Angeles, United States; Shopify for product, cart, order and checkout functions; and Stripe for payments, subscriptions, billing and refunds. Email delivery providers and our limited internal support and fulfillment team process order communications, confirmation photographs and replies, and relevant support messages, contact details and attachments. Logistics providers receive the contact, delivery and order details needed to fulfill an order. These roles do not give every recipient full conversations, journals or reports. Annex A describes their roles and data categories.
7.3 Internal human access
Access to relevant records is currently managed by a limited internal team. Records may be reviewed for customer support, troubleshooting, security, operations and authorized internal AI training or annotation. Operational access is not itself training authorization. SoulMatch has no continuous human crisis monitoring.
7.4 Private Content protections
We do not sell Private Content or disclose or use it for targeted advertising, public testimonials, promotional stories or unrelated third-party marketing. This protection includes chats, AI replies, scans, mood selections, journals, reports, sensitive feedback and their summaries or inferences, including copies in logs. Advertising pixels and advertising-related events must not receive Private Content or sensitive wellness signals. Promotional screenshots are limited to internally created demonstration material, not users' private generated content.
8 Cookies advertising and processing locations
8.1 Essential storage
We use browser storage and session technologies for sign-in, language, service acknowledgements, consent choices, chat continuation, deletion status and relevant local content or preferences. Clearing browser data may remove local access to saved history or session-based records. It does not automatically remove records held on our servers.
8.2 Optional usage statistics
First-party usage statistics are off by default. If you actively allow optional usage statistics using a random session identifier, SoulMatch generates a random session ID for the current browser session and records only allowlisted events: page_view, link_click and button_click. Page paths do not include query parameters or fragments. Click events are recorded only when developers explicitly mark a safe event name. Logs do not include page titles, raw button or link text, element IDs, roles, arbitrary metadata, mood text, chat content or other free text. Usage-statistics logs are kept for 90 days by default. If GPC is detected, usage statistics are not enabled.
8.3 Advertising and your choices
After you consent to advertising tracking, Google Ads may measure eligible purchases only on the bracelet payment-success pages on soulmatch.store and www.soulmatch.store. We isolate this measurement from the application page and do not load advertising code on chat, scan, report, journal or other private pages. The information supplied for a conversion is limited to the conversion destination, transaction identifier, available amount and currency, and a fixed purchase-confirmation page address and title. We do not supply the actual query string, fragment or referring-page address. Google may process technical information needed for the request, such as IP address and device information. These controls do not authorize disclosure of a transaction that reveals sensitive wellness information; the restrictions below continue to apply.
We do not send Private Content or sensitive wellness signals to advertising services. This restriction covers identifiers and page paths that reveal sensitive activity, not just message text. Advertising measurement must not connect private chats, scans, reports or journals to marketing profiles. You can refuse or withdraw advertising permission in privacy settings. Where GPC is detected or advertising is refused, Google Ads is not loaded. Refusal does not prevent ordinary AI conversation.
8.4 Processing locations
Soul Alchemy World Limited is headquartered in Hong Kong, and members of our internal team who administer the service may access relevant records from Hong Kong. The SoulMatch application server is hosted with UCloud in Los Angeles, United States. AI generation uses the Google Gemini API; because Gemini is a globally distributed service, Google processing, caching, abuse monitoring and support may occur in a range of locations used by Google and its affiliates or subprocessors under applicable terms, and may not be limited to the United States. Shopify, Stripe, email-delivery and other providers process information according to their own infrastructure and terms. For users in the United States, this means that personal information you provide — including Private Content sent to the AI for a response — may be processed or accessible outside the United States, including in Hong Kong and in other locations where Google or our other processors operate.
Information may be transferred outside your place of residence, including to Hong Kong where our company is based and to other locations used by our AI and other processors, and may be accessible to authorities under laws in the processing location. For United States users specifically, hosting the application server in the United States does not mean that all processing, support access or provider records remain only in the United States; some processing occurs in Hong Kong and in Google's distributed infrastructure. We remain responsible for our own processing and use appropriate contractual and organizational protections for processors.
8.5 Limits on disclosure and linking
Private Content and sensitive wellness inputs must not appear in advertising parameters, email subjects, notification previews or other unnecessary disclosures. Operational identifiers may be used only for the service, necessary troubleshooting, security, fraud prevention, rights requests or user-requested continuation. They do not create an exception allowing advertising platforms or unrelated vendors to connect Private Content to named accounts, payment records, Shopify profiles or marketing profiles. Hashing does not remove this restriction.
9 Consumer health data
9.1 Categories sources purposes and recipients
This Section 9 is a standalone consumer health data section. It is provided to meet the requirements of the Washington My Health My Data Act and comparable state consumer-health-data laws (such as the Nevada consumer health data law and the health-data provisions of other state privacy laws), and it governs how we handle consumer health data in addition to the general terms of this Policy. Consumer health data or sensitive wellness information may include emotional or mental-health descriptions, mood selections, scan and reflection answers, journal entries, conversations and replies, reports, summaries, safety-related inferences and information revealing that someone seeks emotional support, including relevant product choices or purchases. It is processed for the requested services, related storage, necessary and legally permitted safety and support, and legal obligations. Relevant recipients include AI and hosting/storage processors, necessary support providers, payment/commerce/logistics providers for transaction and fulfillment data, and lawful recipients under legal requirements. Payment, commerce and logistics providers do not thereby receive full chats, journals or reports.
Sources include information you enter, AI outputs and inferences derived from it, service activity and relevant transaction information. AI processors receive the input and context needed to generate an output; hosting processors store the relevant service records; support personnel or providers receive information needed for a specific request; payment, commerce and delivery providers receive necessary transaction and fulfillment information. Identifiable Private Content used for internal development remains subject to Section 4.3.
9.2 Consent and use restrictions
We collect consumer health data only as necessary to provide your requested service or with consent required by law. Sharing beyond applicable processor or requested-service exceptions requires separate consent where required. We do not sell consumer health data or other Private Content. We do not use healthcare-location geofences. Additional categories or purposes require prior updated notice and consent where required.
9.3 Health data rights
Where applicable, you may confirm relevant collection, sharing or sale, access your consumer health data and the required recipient list, withdraw consent and request deletion. Email support@soulalchemyworld.com with the subject line Consumer Health Data Request. We respond within applicable legal periods and provide appeal routes where required. If we deny a request and an appeal right applies, you may reply to our decision or email the same address with the subject Privacy Appeal. We explain the decision, the available appeal procedure and applicable regulator complaint route.
10 Your privacy rights
10.1 Requests and response periods
Depending on applicable law, you may have rights to access, correction, deletion, a copy or portability, withdrawal of consent, restriction, objection and appeal. Contact support@soulalchemyworld.com without sending unnecessary identification or sensitive chat content. We respond as soon as reasonably practicable and within the applicable period: generally 40 days for covered Hong Kong access or correction requests, 30 days for access under Canada's PIPEDA with permitted extensions, and 45 days for covered US state requests with an additional 45 days only where allowed and explained. Where mainland China law applies, our retained service target is a response within 15 working days, subject to any shorter mandatory requirement.
These general response periods do not postpone an advertising opt-out or another choice that must take effect sooner. Such choices are processed within their applicable periods. Any permitted extension is explained with its reason and expected response date. An account-closure waiting period does not pause a privacy request.
10.2 Verification representatives and closure
We may request information reasonably necessary to verify the request or an authorized representative. If a session cannot be linked to you, we may need session details or other proportionate evidence. Chat-history deletion does not itself close the account. Account closure is distinct from cancellation of a separately billed subscription, membership, order or payment obligation.
10.3 United States and California rights
State rights vary and coverage conditions matter. Applicable rights can include rights concerning sale, sharing, targeted advertising, sensitive information and certain profiling. If the California Consumer Privacy Act, as amended, applies to us, California residents may request to know, access, delete, correct, opt out of sale or sharing, limit use of sensitive personal information, and not be discriminated against for exercising those rights.
Use the advertising controls in privacy settings or a supported Global Privacy Control signal to opt out of optional advertising tracking; you may also contact us for rights requests. We do not sell Private Content. Where an optional advertising disclosure is legally considered a sale or sharing of other personal information, applicable opt-out rights apply. Necessary processing and any required sensitive-data consent remain separately governed by this Policy.
Sensitive personal information. Under the California Consumer Privacy Act as amended by the California Privacy Rights Act, and comparable state privacy laws, certain information we process may be sensitive personal information (SPI). For SoulMatch this can include your full birthday, your mood and emotional data, and wellness inferences we or the AI derive from your Private Content, as well as other information that reveals your emotional or mental-health state. Where applicable law gives you the right to limit the use and disclosure of sensitive personal information to what is necessary to provide the service you requested, you may exercise that right by using the controls in privacy settings or by contacting support@soulalchemyworld.com. We do not use SPI to infer characteristics about you for unrelated purposes, and we do not sell or share SPI or other Private Content for cross-context behavioral advertising.
11 Security updates and contact
11.1 Security
We apply safeguards appropriate to information sensitivity and applicable law. No system is perfectly secure. If a data breach affecting your personal information occurs, we will notify affected users without undue delay and, in any event, consistent with the deadlines required by applicable law. Notice will ordinarily be sent to the email address on file for your account and will describe, to the extent known and permitted, the nature of the breach, the categories of information involved and the steps you can take. Because the service involves consumer health and sensitive wellness data, a breach involving that data may trigger faster or additional notification obligations (including to regulators), and we will comply with those obligations. We also notify authorities when applicable legal thresholds and deadlines require.
11.2 Updates
We publish an effective date and give appropriate notice of material changes. We do not silently expand the use of previously collected Private Content. Where required, we obtain a new choice before a changed purpose begins.
11.3 Contact and complaints
For questions, requests or complaints, contact support@soulalchemyworld.com. You may also complain to a competent regulator, including the Hong Kong Office of the Privacy Commissioner for Personal Data, the Office of the Privacy Commissioner of Canada or an applicable provincial authority, the California Privacy Protection Agency or your state attorney general, and the Cyberspace Administration of China where applicable. This Policy does not restrict any mandatory local complaint or appeal right.
A Service providers
A.1 Google Gemini
Receives relevant prompts, context and outputs for AI generation. Google processing and support may occur in locations used by Google and its subprocessors. Sections 4.1, 4.2 and 4.4 explain caching, 55-day abuse-monitoring records, limited human review, policy-enforcement use and restrictions on general product improvement.
A.2 UCloud
Hosts the SoulMatch application and relevant service records in Los Angeles, United States. Our current backup position is described in Section 6.5.
A.3 Shopify and delivery providers
Shopify supports product, cart, order, checkout and related product or order synchronization using relevant customer, product, delivery and transaction data. Delivery providers receive necessary order and contact details for fulfillment. Their infrastructure may involve processing outside your residence. Private Content is not supplied for advertising or merely because a transaction occurs.
A.4 Stripe
Stripe processes payment and transaction information for payments, subscriptions, billing and refunds under its applicable infrastructure and terms. For a customized bracelet checkout, this includes the necessary customer and delivery information, order or design identifiers, price and currency, and order metadata such as the selected bead layout, wrist measurement, gift message, confirmation email and purchase-notice acceptance details. We do not place birthday, mood, intention or private return-page context in the success or cancellation URLs sent to Stripe; a design identifier lets SoulMatch restore the saved design separately. SoulMatch receives information needed to administer the transaction; it does not thereby store full payment-card details.
A.5 Email delivery and support
Email delivery providers and our limited internal support and fulfillment team handle contact details, order information, messages and relevant attachments. An internal fulfillment notification may contain the order and user identifiers, confirmation email, delivery name, address and phone number, design image, bead layout, gift message and transaction amount. We send actual bracelet photographs to your confirmation email before dispatch and process your reply to confirm shipment or resolve a concern. Order correspondence is not an authorization for marketing or training, and this role does not give recipients access to all conversations. Please avoid unnecessary sensitive content in email subjects or gift messages.
A.6 Advertising and usage statistics
Google Ads receives only permitted purchase-conversion information after advertising consent, through isolated measurement on the designated payment-success pages. Private pages and their context are excluded. First-party statistics are separate, default-off and use minimized session events. Section 8 and Annex B explain choices, GPC and the prohibition on advertising use of Private Content.
B Browser storage and tracking
B.1 Essential storage
Used for sign-in, language, service acknowledgements, chat continuation, deletion status and relevant local content or preferences. Clearing browser data removes local access, not all backend information. This is not training consent and does not imply a separate memory editor.
Bracelet design history, composition snapshots and pending account-synchronization records may also be stored in your browser to restore your work. They can contain the birthday and personalization context described in Section 2.6. Clearing browser storage does not automatically delete account-synchronized designs, paid orders or order emails.
B.2 Optional usage statistics
First-party statistics are default-off, consent-based, allowlisted and minimized. They use random current-session identifiers, path-only page views and developer-marked safe event names. They do not collect query parameters, titles, raw button/link text, arbitrary metadata, mood text, chat content or free text. Retention is 90 days by default, configurable from 1 to 365 days. If GPC is detected, statistics do not start.
B.3 Advertising technologies
Google Ads is gated by advertising consent and limited to designated bracelet payment-success pages. We do not send emotional-support subscription conversion events to advertising platforms. Measurement runs in an isolated frame with access only to the explicitly supplied conversion fields described in Section 8.3; it does not receive the parent page content, its query parameters or its referrer. Leaving the permitted page or withdrawing consent stops the active measurement frame. Google may process the request-related technical information described in Section 8.3. Any optional advertising technology within a Shopify purchase flow is subject to the applicable advertising choice and required opt-out protections. Using Shopify for checkout does not itself mean consent to optional advertising. The Private Content restrictions apply across both flows.