Trust Centre
On this page
Manufacturers trust BrandedMark with their product data and their relationships with their customers. This page explains, plainly, how we protect that data and where it lives. It describes how BrandedMark works today. Anything marked Planned is not in place yet, and we update this page when it is.
Contacts. Security: [email protected]. Privacy and data protection: [email protected]. Contracts and DPAs: [email protected].
Company. Branded Mark Limited, registered in England and Wales, company number 14452041. Registered office: Broomfield House, Blakedown, Kidderminster, DY10 3LA, United Kingdom. VAT number: GB 530 5138 24. ICO registration: ZB954818.
Compliance status
| Area | Status as at 1 November 2026 |
|---|---|
| UK GDPR and EU GDPR | Data Processing Agreement included in our terms (DPA); public sub-processor list |
| ICO registration | Registered, ZB954818 |
| EU representative (Art. 27 GDPR) | DataRep: [email protected] (quote "BrandedMark"); see our Privacy Policy |
| EU legal representative (Digital Services Act, Art. 13) | DataRep: [email protected] (quote "BrandedMark"); see our Acceptable Use Policy |
| EU data residency | Customer data hosted in Microsoft Azure in the EU (Netherlands, with a copy in Ireland). AI requests may be processed outside the EU: see AI governance |
| EU AI Act | Transparency obligations (Art 50) for consumer-facing AI: see our AI Statement |
| EU Data Act | Switching and data-export rights built into our Customer Terms |
| Cyber Essentials (UK) | Planned |
| Independent penetration test | Planned |
| ISO/IEC 27001 | Not certified. We are aligning our controls with ISO 27001:2022. |
| SOC 2 | Not held. Our hosting provider, Microsoft Azure, holds ISO 27001, ISO 27701 and SOC 2 Type II reports. These cover Microsoft's controls, not ours. |
Infrastructure and data residency
- Cloud provider. Microsoft Azure.
- Region. Customer data is hosted in the EU (Netherlands), with a geo-redundant copy of files in Ireland.
- EU regions only. Our production hosting is restricted to approved EU regions. AI requests and a few global suppliers are described under AI governance and in the sub-processor list.
- Separate environments. Production is isolated from development and testing, and protected against accidental deletion.
Encryption
- In transit. TLS 1.2 or higher on all public connections to our services, with HSTS enforced.
- At rest. AES-256 encryption for the database, file storage, managed disks and backups, with Microsoft-managed keys.
- Keys and secrets are held in a dedicated secrets vault, never in code.
Access control
- No passwords to our data stores. Our services reach our data stores without passwords or keys, and deployments use short-lived credentials.
- Need to know. Access to our production cloud platform is limited to our directors, an emergency account and our own automated tools, through our central identity provider. Our automated tools work under the same confidentiality and security controls as our staff, and only to operate, support and secure the service.
- Your users. Your admins control roles and access scopes for each user and brand, and can require two-factor authentication or passkeys for all staff in your account.
- Your consumers sign in without a password, using a one-time code or a passkey.
Application security
- Strict Content Security Policy, secure host-only cookies and anti-forgery protection.
- Rate limiting to protect against abuse.
- Malware scanning of uploaded files.
- Short-lived signed links for private files.
- Signed webhooks with server-side request forgery protection. Inbound webhooks from our providers are signature-verified.
- API keys are stored only as hashes and scoped to the public API.
- Code changes go through pull requests with automated tests.
- Every build checks our dependencies for known vulnerabilities.
- Internal security reviews of the platform.
Tenant isolation
BrandedMark is multi-tenant:
- Every record belongs to one customer (tenant), and access is enforced by tenant-scoped checks in the application, covered by automated tests.
- Each customer's files are stored separately.
- Each customer's integration secrets are stored in our secrets vault and used only for that customer.
- What is public. Images, videos and branding you upload to your media library, and the QR-code images we generate, are stored so that anyone with the exact link can view them, because they appear on your product pages. The links are long random identifiers, and the storage can't be browsed or listed. Don't put anything confidential in the media library. Everything else is private: product and passport documents (including public ones, which we serve through the app), consumer uploads such as receipts and photos, support attachments, exports and print files. It is only reachable through the app or short-lived signed links.
Backups and resilience
| Area | Today |
|---|---|
| Backups | Continuous database backup and versioned file storage, with a geo-redundant copy of files in a second EU region (Ireland). Deleted data rolls out of backups within 65 days (stored integration credentials within 90 days); see the DPA, clause 11.3 |
| Recovery | A documented disaster-recovery plan, with recovery time and recovery point targets for a restore within the primary region; see the DPA, Annex II |
| Status page | status.brandedmark.com: live status, uptime history, incidents and scheduled maintenance |
Logging and monitoring
- We monitor the service, with alerting.
- Your account has an audit log of significant actions, kept for 60 days.
People and processes
- Everyone we authorise to access customer data is bound by confidentiality obligations.
- Access is granted on a need-to-know basis and removed when someone leaves.
- We have a written incident response process.
Incident response
If a personal data breach affects your data, we tell you without undue delay, aiming for 48 hours and in any event within 72 hours of becoming aware. We then give you the information you need to meet your own obligations. See the DPA, clause 8. Report security issues to [email protected].
AI governance
- AI features run on Google Gemini models through the Gemini API. Google doesn't commit to a processing location for this API, so AI processing may happen outside the UK and EU (see the AI Statement).
- Our AI provider never trains on your data, and we don't train AI models shared between customers on it.
- Your assistant keeps short summaries of well-rated past answers and reuses them only in your account, to answer similar questions about the same product.
- Separately, unless it's switched off, we improve our AI features from support questions, answers and satisfaction ratings, after removing personal details and your company and brand names (see the FAQ below).
- Features we label Preview are optional and come without the SLA.
- AI answers routine questions and drafts content within rules you set. Your team can review, take over or switch it off.
See our AI Statement, which also describes how people are told when they're dealing with AI, and the FAQ below.
Resources
- Data Processing Agreement
- Sub-processors
- Privacy Policy
- Customer Terms
- Service Level & Support Policy
- Acceptable Use Policy
- AI Statement
- Responsible Disclosure and security.txt
- Status page
- On request, under NDA (email [email protected]): answers to your security questionnaire, and a summary of our architecture and disaster-recovery plan.
Frequently asked questions
Security information
What certifications or accreditations does BrandedMark hold?
We don't hold an independent certification. Cyber Essentials is planned, and we're aligning our controls with ISO 27001:2022. Our hosting provider, Microsoft Azure, is certified to ISO 27001, ISO 27701 and SOC 2 Type II; those certifications cover Microsoft's controls, not ours.
Do you have a documented information security policy?
Yes. It's owned at board level and reviewed at least once a year. It covers access control, secure development, incident response, supplier management and acceptable use.
Is all client–server communication encrypted?
Yes. Public connections to our services use TLS 1.2 or higher, and HSTS is enforced.
Is data encrypted at rest?
Yes. The database, file storage, managed disks and backups use AES-256 encryption, with Microsoft-managed keys.
Do you carry out penetration tests?
We carry out internal security reviews. An independent penetration test is planned; once it's done we can share a summary under NDA.
What security training do staff receive?
Our staff training covers security, data protection and the responsible use of AI, with a refresher every year.
Do you have cyber insurance?
Our Technology Professional Indemnity and Cyber insurance is being provisioned. Ask us for current details and certificates.
Do you use open-source software?
Yes. We use well-maintained open-source components. We monitor them for vulnerabilities and update them through our normal development process.
Product security
Is two-factor authentication available?
Yes, for your staff: authenticator-app codes (TOTP) and passkeys. Your admins can make two-factor or passkeys mandatory for everyone in your account. It's off by default, and we recommend turning it on.
Do you support passkeys?
Yes, for both staff and consumers.
Do you support single sign-on (SSO)?
No. Tell us if you need it.
Where can I see uptime and incidents?
On our status page, status.brandedmark.com, which shows daily uptime and incident history. We aim for 99.5% monthly availability. Our standard terms don't include service credits, but if availability falls below 99.0% three months in a row you can leave with a pro-rata refund. Details are in the Service Level & Support Policy.
Personal data
What personal data does BrandedMark process?
It depends on the features you use:
- Your staff: name, email, role, and sign-in and security records.
- Your consumers: contact details, product registrations and proof of purchase, serial-label photos, cases and returns, orders, conversations, marketing preferences and scan events. Also IP address and the country derived from it, fraud-prevention signals and, where the consumer allows it in their browser, precise location at registration.
- Your business customers' contacts.
The full list is in Annex I of the DPA.
Who are the data subjects?
Your staff users, your end consumers and product owners, and contacts at your business customers, dealers and installers.
Is BrandedMark a controller or a processor?
For the data you put into BrandedMark, including about your consumers, business customers and staff, you're the controller and we're your processor. We're a controller only for our own business needs: managing your account and billing, your staff users' sign-in and security records, securing the platform, and meeting our legal obligations. Usage statistics we make from your account are made on your instructions, as your processor. See the DPA, clause 2.
Do you process special category (sensitive) data?
Not by design. The platform isn't built to collect it, and our terms ask you not to configure it to. Free-text messages and uploads may occasionally contain it if a consumer chooses to include it.
How and where is data stored?
In Microsoft Azure in the EU, encrypted at rest. The primary region is the Netherlands, with a geo-redundant copy in Ireland.
Can I choose a different hosting region?
No. All customers are hosted in the EU. Talk to us if you need another region.
Where is your DPA?
It's built into our Customer Terms and published at /legal/dpa. You don't need to sign it separately, but we're happy to countersign a copy.
Do you use sub-processors?
Yes, a short list, published at /legal/subprocessors. We give 30 days' notice of changes.
Jurisdiction and data transfer
Where is BrandedMark registered?
In England and Wales, company number 14452041. Registered office: Broomfield House, Blakedown, Kidderminster, DY10 3LA. VAT number: GB 530 5138 24.
Does data leave the UK and EU?
Customer data is hosted in the EU. Our team is in the UK; the EU recognises the UK as adequate, so EU data can flow to it. Some sub-processors process data outside the UK and EU: Google's Gemini API for AI features (no committed processing location), Cloudflare's global network, and Stripe. These transfers are protected by Standard Contractual Clauses with the UK Addendum and, where the provider is certified, the EU-US Data Privacy Framework. If you use address lookup or maps, Google Maps Platform works under Google's own terms. Our sub-processor page lists each supplier and where it operates.
Have you ever received a government request for customer data?
No, as at 20 October 2026. If we do, we'll tell you unless the law forbids it, challenge it where there are grounds to, and disclose only the minimum. See DPA clause 13.
Data access
Who at BrandedMark can access our data?
Only named BrandedMark staff, when needed to provide support, fix problems or keep the service secure. We also use our own automated tools to operate and support the service; they work under the same confidentiality and security controls as our staff. Access to our cloud platform is through our central identity provider.
Can BrandedMark staff access our data or act as our users?
Our support team may view your tenant read-only, including as a named user would see it, to help with an issue or investigate an incident. Our automated tools are used only to operate, support and secure the service, and never view your tenant as a user.
- Read-only. We only act as a user or make changes when you ask us in writing, for example in a support ticket, or one of your administrators approves it.
- Every access is recorded. We record each time our staff access your tenant, who did it, when and why, with the ticket or incident reference, and keep that record for at least a year. An administrator can ask us for a copy at any time.
- You can ask us to stop. Some issues may then take longer to resolve, or we may not be able to resolve them. See Customer Terms clause 18.5.
Do third parties have access to our data?
Only the sub-processors on our list, under written data protection terms, and Google Maps Platform, under Google's own terms, if you use address lookup or maps. Services you connect yourself, such as your ERP, carrier or webhooks, receive data on your instructions.
Which critical functions are outsourced?
Hosting, network security, email delivery and AI models, plus payments and text messages when you use those features. Each provider is listed on our sub-processor page. We build and operate the platform ourselves.
Is access logged?
Actions in your account are recorded in an audit log, kept for 60 days. Our sign-ins to the cloud platform are logged by our identity provider. Each support access by our staff is recorded and the record kept for at least a year.
Data protection measures
Do you have a privacy policy?
Yes: /legal/privacy. We also provide a default consumer privacy notice that you can publish on your BrandedMark portal and edit.
Do you have a Data Protection Officer?
We're not required to appoint a DPO. Data protection is owned at board level, and you can reach us at [email protected].
Are you registered with the ICO?
Yes: ZB954818.
Do you have an incident response process and a breach register?
Yes. We keep a register of security incidents, and breach notification works as described under Incident response above.
Data retention
How long do you keep data?
For as long as you're a customer, subject to automatic clean-up rules:
- conversations, with their AI traces: 2 years after the last activity;
- AI answer summaries: until your account is deleted, or earlier if you ask;
- scan events: 90 days;
- your account's audit log: 60 days;
- operational logs: 30 days;
- product recall notification records: 10 years;
- deleted items: permanently removed after 30 days.
Our own account and billing records are kept as described in our Privacy Policy. Financial records are kept for 6 years.
Can we delete data at any time?
Yes. Your admins can delete records and handle consumer erasure requests in the product, with an optional review delay. Consumer erasure doesn't remove AI processing records or AI answer summaries; on your request we remove a named consumer's personal data from them within 5 business days. You can also ask us at [email protected].
What happens when we leave?
- You have at least 30 days after the end of your agreement to take your data, free of charge, using the in-product exports and the API.
- An administrator can also ask support for a complete export, as Customer Terms Schedule A describes.
- After that we delete your data within 30 days.
- Deleted data rolls out of backups within a further 65 days, and stored integration credentials within 90 days. See Customer Terms Schedule A.
AI at BrandedMark
Does BrandedMark use AI?
Yes, to help your team and your customers. That means:
- an AI support assistant for consumer chat and email;
- reply drafting for your agents;
- deciding when to hand a conversation to your team;
- remembering well-rated answers within your account;
- reading serial-number labels from photos;
- summarising and quality-scoring conversations;
- drafting support playbooks and procedures for your team;
- generating product experience content and media descriptions;
- searching your product documents.
Which providers power BrandedMark's AI?
Google Gemini models, accessed through the Gemini API. Google doesn't commit to a processing location for it, and doesn't use our data to train its models or improve its products. We don't train foundation models.
Is our data used to train AI models?
Not by our AI provider: under its terms, Google doesn't use your prompts or responses to train its models or improve its products.
- Answer memory: as part of the assistant itself, it keeps short summaries of the issue and the fix from AI-answered conversations your consumers rate highly, and reuses them to answer similar questions about the same product, only in your account. These summaries may contain personal data from the original conversation. See DPA clause 3.2(g).
- Our own improvement work: separately, unless it's switched off, we learn from support conversations (the question, the answer and the satisfaction rating) to make answers more accurate, for example which spare part fits or what an error code means.
- What's removed first: names, contact details and other personal data, and your company and brand names. Personal data never goes into any training or evaluation set.
- Your product knowledge: what we learn from your product information (models, error codes, part numbers, your documents) improves only your own assistant.
- For everyone: we also improve the instructions, guardrails and checks all customers share. Those never contain your content, and we don't train shared AI models on your data.
- On or off: it's on for accounts created after our Customer Terms take effect, and off for earlier accounts unless they switch it on.
- Switching it off: email [email protected], or set it in your Order Form. See Customer Terms clause 6.4.
Do AI providers store our data?
Google keeps prompts and responses for a limited period (currently up to 55 days) to detect and prevent misuse of its service, and items its systems flag may be reviewed by people. Your persistent data stays in our Azure EU environment, apart from document-search indexes of product documents you mark Public and Include in AI, which Google holds (location not committed) until you remove them. We keep a trace of AI prompts and responses with your conversation records, so you can review what the AI did. These traces follow the same retention rules as the conversations.
What goes into and comes out of the AI?
- In: messages, the case, order and return details needed to answer, serial-label photos, and your product documents.
- Out: suggested or sent replies, extracted serial numbers, summaries and scores, and drafted content.
The AI doesn't process biometric data. Our built-in AI tools can look up information, open cases and send messages; they can't approve or refuse warranty claims, returns or refunds. If you connect your own tools to the assistant, you decide whether they may act without human review.
Who stays in control?
You do:
- You configure what the assistant may do and which tools it can use.
- Your agents can review, edit or take over any conversation.
- AI-generated pages and text in the designer stay drafts until you publish them.
Can we turn AI off?
Yes. You choose which AI features you use. The AI support assistant in product chat and AI email replies are on by default once you turn on chat or support email, and you can switch each off in settings. We'll switch AI off for your whole account on request.
Are consumers told when they're talking to AI?
Our AI Statement describes how people are told when they're dealing with AI, in the chat and in AI-written emails, and how AI-generated content is marked.
Is BrandedMark's AI "high-risk" under the EU AI Act?
No. Our AI features support customer service and content creation for physical products. They aren't used for any Annex III high-risk purpose (such as employment, credit, education, essential services or biometrics), and they aren't safety components of regulated products. They're subject to the Act's transparency obligations, which fall on us as provider. Our AI Statement describes the measures.
How do you manage AI accuracy?
- The assistant answers from your own product information and documents.
- Where you've set up handover to your team, it hands over when it's unsure or when your rules require it.
- Conversations can be quality-scored and reviewed.
- AI can be wrong. We recommend you keep a person in charge of anything consequential, such as refunds or warranty decisions. Our built-in AI tools can't make those decisions.
Who owns AI outputs?
As between you and us, AI outputs generated for your account are part of your Customer Data, and we claim no rights in them.
Can we choose the model?
No. We select and govern the models centrally, and test a new model on our own features before we use it.
What are Preview AI features?
A feature is a Preview feature only when we label it Preview (or beta) in the product. Preview features are off until an admin switches them on, one by one or by choosing automatic enrolment. They're outside the SLA and provided as is. We tell you about any charge before you switch one on.
What protects the AI from misuse?
Our terms and Acceptable Use Policy prohibit:
- jailbreaking and prompt injection, including instructions hidden in uploaded files, web pages, emails or webhooks;
- attempts to reveal system prompts;
- using AI tools to reconstruct or clone the platform;
- systematically capturing how our AI features behave to build a competing model. The assistant also only uses the tools and data you enable.
Digital Product Passports
What happens to our product passports if we leave?
- You can export the passport data and GS1 Digital Link mappings.
- If you ask before your data is deleted, we'll redirect your links to a new host for 12 months at no charge.
- Ask us about Link Continuity, a paid option that keeps your links and passports live, read-only, after you leave.
- Otherwise, once your data is deleted, your links show a notice that the product link is no longer active.
See Customer Terms Schedule B.
Is BrandedMark a certified DPP service provider?
No. Certification for DPP service providers under the EU Ecodesign Regulation hasn't been defined yet. We implement GS1 Digital Link and design to the DPP requirements published so far (Regulation (EU) 2024/1781), and we intend to seek certification as soon as a scheme is available.
ESG
What's your approach to ESG?
Our platform supports product registration, repair, returns and product passports, which can help products stay in use for longer. We run on Microsoft Azure, which publishes its own sustainability commitments.
Updates
| Date | Category | Update |
|---|---|---|
| 20 October 2026 | Trust | Trust Centre launched |
Version history
| Version | Date | Change |
|---|---|---|
| 1.2 | 1 November 2026 | EU representatives under the GDPR and the Digital Services Act. |
| 1.1 | 20 October 2026 | Added our VAT number. |
| 1.0 | 20 October 2026 | First published version. |