Privacy Policy
Effective date: 1 October 2026 | Version: 1.0
This Privacy Notice is the global privacy baseline for PINKY. Additional or different requirements may apply depending on where you are located or where we offer Services. A jurisdiction-specific supplement adds to this Notice; it does not silently replace the baseline. The Japan Supplement applies when the Service treats you as Japan-facing under the applicable routing policy, regardless of whether you use the Japanese or English interface.
1. Controller
UNISON-CHORD LTD is the controller of personal data processed for the purposes described in this Notice unless a product-specific notice says otherwise. Contact: info@pinkycats.com; 9 Caxton House, Broad Street, Cambourne, Cambridge, United Kingdom, CB23 6JN. Company number: 17405540.
2. Scope
This Notice covers PINKY websites, PINKY CATs, PINKY accounts, assessments, saved results, feedback, communications, personalisation, research and product-improvement activities, and future PINKY Services that refer to this Notice. A future product may provide additional privacy information where its processing materially differs.
3. Information you provide
Depending on the feature, we may collect age confirmation; email address and other account/contact details; authentication and recovery information; assessment answers; the person or scenario you had in mind where a question requires context; feedback; support messages; preferences; communication choices; purchase information; user-provided conversation or memory content; and information you deliberately submit through future features.
4. Assessment, result and derived information
We process assessment attempt identifiers, question/assessment versions, answers, scores, type mappings, result content, completion state, timestamps and result history where saved. We may derive tendencies, profiles, classifications, compatibility or preference signals, psychometric measures, longitudinal changes and other inferences from saved information. If an inference intentionally concerns sex life, sexual orientation, health or another UK GDPR special category, we treat it as special-category data where Article 9 applies.
5. Technical, security and usage information
We may process IP address where necessary for security, fraud prevention, routing or delivery; device/browser and operating-system information; timestamps; request and event identifiers; authentication/session events; consent and policy evidence; service errors; security events; and limited product-interaction data. We design logging so raw assessment answers, sensitive results, full authentication tokens and unnecessary direct identifiers are not written to logs or telemetry.
6. Pre-registration CATs answers
Before registration/account binding in the initial PINKY CATs flow, raw assessment answers are processed transiently for active progression, scoring and result generation and are not durably persisted. Session expiry or abandonment removes transient raw-answer state. Our design requires logs, analytics, crash reporting, traces, backups and vendor telemetry not to create a hidden persistent copy of those raw answers. We may retain separate evidence of the consent event and minimal technical/security events where necessary.
7. Account binding and saved data
If you register and bind the assessment to a verified account, the applicable assessment data and result may then be persistently saved. The binding event does not overwrite the earlier consent-event history. We keep the PINKY account identity/authentication layer logically separated from sensitive assessment/psychometric data and use controlled linkage rather than treating the email address as the assessment record key.
8. Save & Resume
Where offered, Save & Resume after authentication may persist an incomplete assessment for a limited operational period, currently designed as up to 30 days for the initial release. Save & Resume is service delivery, not a research, marketing or privacy consent. An incomplete attempt may be deleted or expire after the configured period.
9. Special-category data
PINKY CATs intentionally processes information concerning sex life and sexual orientation. Under UK GDPR these are special categories of personal data. Depending on future features, health or other special-category data may also arise. We identify both an Article 6 lawful basis and an Article 9 condition before processing special-category data. For the current CATs assessment, our Article 9 drafting position is explicit consent under Article 9(2)(a), obtained separately before the assessment begins.
10. Heightened-sensitivity information
Some information may be deeply private even where it does not fall within a statutory special category. We apply heightened safeguards to relationship information, intimate preferences, psychological or wellbeing-related information, conversational memory, longitudinal profiles and similar sensitive context based on risk as well as formal legal classification.
11. Purposes
We use personal data to provide assessments and results; create and secure accounts; save history; personalise Services; provide continuity across enabled features; deliver service communications and user-requested result notifications; enable user-directed sharing; process payments; support users; prevent abuse and protect security; evaluate, validate and improve assessments and Services; conduct internal research and longitudinal analysis; develop psychometric methods, algorithms, models and PINKY Intelligence; operate optional external research where separately authorised; comply with law; and establish, exercise or defend legal claims.
12. Article 6 lawful bases
We use contract where processing is objectively necessary to provide a Service you request or manage the account/transaction; consent where the law or our chosen design requires a genuine optional choice, including direct electronic marketing where applicable; legitimate interests where we have a proportionate interest such as service security, fraud prevention, limited internal evaluation or product improvement and the balancing test supports it; and legal obligation where processing is required by law. A single data item may be processed for different purposes under different bases. We document the purpose-level basis rather than assigning one basis to an entire database.
13. Article 9 explicit consent and scope
For CATs special-category answers, Article 9(2)(a) explicit consent is the current condition for the specified assessment processing. The consent request identifies the sensitive nature of the information and requires an affirmative, unchecked action. The assessment is unavailable without this consent because processing the relevant answers is objectively necessary to provide that assessment. We do not treat this consent as permission for unrelated external research, marketing or materially different future purposes.
14. Post-registration special-category processing
Where saved CATs data remains special-category data, post-registration storage and reuse must remain within a valid Article 6 basis and Article 9 condition. Our drafting position is that Article 9(2)(a) explicit consent may support post-registration persistence, result history, personalisation, internal validation/research and Intelligence/algorithm/model improvement only to the extent those purposes are sufficiently specified in the consent and associated privacy information. We will not silently widen an earlier consent. A materially different purpose requires a new assessment of lawful basis, Article 9 condition, transparency and, where required, renewed explicit consent.
15. Internal service improvement and internal research
We distinguish internal evaluation, validation, quality assurance, psychometric development, longitudinal analysis, algorithm/model improvement and other internal research from optional external research sharing. Declining the external research checkbox does not itself disable internal activities. Each internal purpose must have its own documented Article 6 basis and, where special-category data is processed, an Article 9 condition. We apply minimisation, access controls, research-specific identifiers and de-identification appropriate to the activity.
16. External research
External research participation is optional. We use the separate External Research Participation & Sharing Notice when an approved researcher or institution independently receives, controls or uses individual-level data outside our ordinary controlled processing. We do not treat an institutional affiliation alone as external sharing where the person remains within a controlled UNISON-CHORD purpose and access environment. Aggregate or effectively anonymous publication is assessed separately.
17. Pseudonymisation, anonymisation and aggregation
Pseudonymised data remains personal data where a person can be re-identified using additional information. An irreversible-looking identifier alone does not make a dataset anonymous. Before treating data as effectively anonymised, we assess the dataset and reasonably available means of re-identification, including quasi-identifiers, rare combinations, longitudinal precision, free text, linkage and model leakage. We may use generalisation, suppression, aggregation, sampling, controlled environments or destruction of linkage paths as appropriate.
18. Derived Intelligence, models and commercial use
We may create aggregate statistics, effectively anonymised datasets or analyses, scoring methods, algorithms, models and other derived Intelligence from lawfully processed information. UNISON-CHORD may use, commercialise or license such non-personal or otherwise lawfully processed outputs as company intellectual property. We do not assume an output is non-personal merely because it is derived. Before external licensing or sharing, we assess whether an output can reveal, reconstruct, memorise or permit inference about identifiable individuals, particularly sensitive information. Raw individual assessment responses are not licensed or disclosed externally under the mandatory registration Terms merely because they contributed to development.
19. AI and large-language-model processing
We may use approved external AI/LLM providers to generate or assist with personalised outputs or product development. Our default architecture keeps identity, raw assessment history and long-term memory under our control and sends only context necessary for the task, using pseudonymous or rotating processing identifiers where feasible. We prefer configurations that act on our instructions, do not use PINKY user data for cross-customer model training, provide zero or minimal retention, and support suitable contractual, security and transfer safeguards. Any approved stateful provider feature requires a defined retention/deletion design and an exit or migration path.
20. Memory and conversations
If future Services offer conversation or memory, we may process raw conversation content, extracted memory signals, preferences and longitudinal profiles. Raw conversation retention and longer-lived memory are separate concepts: a memory feature may store selected structured signals without indefinitely retaining every raw conversation. The relevant product notice will explain what is remembered, how to review or delete it, and whether any external AI provider receives context.
21. Service communications
We may send authentication, security, recovery, outage, maintenance, material Terms/Privacy changes and other messages genuinely necessary to operate an existing Service. These are service or administrative communications and are not converted into marketing merely because they use email or another messaging channel. We do not use the service-message category as a route to send promotions.
22. Marketing
General marketing is optional and separated from account creation and the sensitive-data consent. Marketing choices are channel-aware and not pre-ticked. A promotional announcement of a new feature, new assessment, campaign, newsletter or invitation to try a new Service is normally treated as marketing. Where PECR or equivalent rules require consent, we send such messages only with the required permission or another applicable statutory route. Withdrawal does not suppress necessary security or service messages.
23. Special-category personalised marketing
We do not treat general marketing permission as permission to use sex-life, sexual-orientation or other special-category data to decide which marketing an individual receives. If we introduce such profiling, we will perform the required legal and DPIA analysis and obtain a separate explicit consent where required before launch.
24. Recipients and processors
We may use carefully selected providers for cloud hosting, database/infrastructure, email and messaging, payments, security, support, analytics, AI/LLM processing and professional services. We disclose only information reasonably necessary for the relevant role and use contracts, access restrictions and other safeguards appropriate to the provider. We may also disclose information where required by law, to protect rights or safety, or in connection with a corporate transaction subject to applicable safeguards.
25. Production location and GitHub boundary
The primary PINKY production environment is designed for the AWS London region. Some supporting providers may operate outside the UK. GitHub is a US-based development service, but under the current architecture production assessment responses, production identity data and research exports are not stored there. GitHub should not be treated as a recipient of PINKY user assessment data unless implementation changes. Repository metadata or operational personal data, if present, is assessed separately.
26. International transfers
Where UK GDPR restricted-transfer rules apply to a transfer outside the UK, we use an available lawful mechanism such as applicable UK adequacy regulations or appropriate safeguards, including the UK International Data Transfer Agreement or UK Addendum where appropriate, and perform the required data-protection/transfer-risk assessment. We do not use user consent as the routine default mechanism for systematic transfers. Provider location, remote access and onward subprocessors are assessed, not merely the physical server region.
27. Retention
We retain identifiable data only for as long as reasonably necessary for the relevant purpose, taking account of service need, user expectations, sensitivity, legal obligations, security, disputes and research design. Pre-registration raw CATs answers are transient and not durably retained. Authenticated incomplete Save & Resume attempts are designed for a limited period (currently up to 30 days for initial release). Account and saved result data are generally retained while the account or feature remains active and then deleted or de-identified subject to lawful exceptions. Consent/Terms/notice evidence may be retained longer where needed to demonstrate compliance. Security logs and backups use separate operational schedules.
28. Deletion and research assets
When you delete an account or validly request erasure, we remove or de-link identifiable account data subject to lawful retention. Deletion must sever the re-identification path before a research asset is treated as effectively anonymous. Data that has already been rendered effectively anonymous is no longer linked to an identifiable person and cannot ordinarily be retrieved for individual deletion. Pseudonymised data is not treated as anonymous merely to avoid an erasure request.
29. Security
We use technical and organisational measures appropriate to the sensitivity and risk of the data, including encryption in transit and at rest where implemented, logical separation of identity from restricted assessment data, purpose-specific identifiers, least-privilege access, controlled linkage, environment separation, audit logging, key-management controls, restricted human decryption, security testing and incident-response procedures. We do not publish operational cryptographic topology, live keys or details that would weaken security.
30. Your UK data-protection rights
Subject to applicable conditions and exceptions, you may have rights to access personal data, correct inaccurate data, erase data, restrict processing, receive portable data, object to processing based on legitimate interests or direct marketing, and not be subject to certain solely automated decisions with legal or similarly significant effects. Where processing relies on consent, you may withdraw consent at any time without affecting the lawfulness of processing before withdrawal.
31. Withdrawing special-category consent
Withdrawal stops future processing that depends on that consent for the relevant purpose and is propagated to applicable downstream processing. It does not rewrite the historical record that consent was previously given or automatically require deletion of every record where another legal obligation or lawful retention applies. Stopping or abandoning an assessment is not, by itself, a formal withdrawal of consent. We will provide an account/settings or contact route for withdrawal.
32. Automated decision-making and profiling
Assessment scoring and profiling may automatically generate a result or classification. The current consumer assessment is intended to provide informational output rather than a decision that produces legal effects or similarly significant effects about you. If we introduce solely automated decision-making within the scope of applicable statutory restrictions, we will provide the additional information and safeguards required by law.
33. Complaints
Please contact us first at info@pinkycats.com if you have a privacy concern. You also have the right to complain to the UK Information Commissioner where UK data-protection law applies. Japan-facing users may also have rights and complaint routes under Japanese law as described in the Japan Supplement.
34. Changes to this Notice
We version this Notice. Material changes are presented in an appropriate way before affected processing where required. A new Notice version does not silently expand the scope of an old consent. Where a new or materially different purpose requires renewed consent, we ask for it separately.
35. Contact
UNISON-CHORD LTD, 9 Caxton House, Broad Street, Cambourne, Cambridge, United Kingdom, CB23 6JN. Email: info@pinkycats.com. Company number: 17405540.
Effective date: 1 October 2026.