How We Handle Your Data
Client data notice and sub-processor list
Last updated: 26/10/2026
This notice explains what data we collect during an audit engagement, who processes it, where it is stored, and how long we keep it. It supplements our Privacy Policy and the Data Processing Agreement we sign with every client.
What we collect during an engagement
Kick-off questionnaire: your services, brand names, competitor names, and details of key personnel (names, roles, credentials, LinkedIn URLs) that you choose to share.
Website data: a crawl of your publicly accessible website content and technical configuration.
Google Search Console and Bing Webmaster Tools: read-only access you grant us, used to analyse search queries and page performance. We never modify your properties, and access is revoked when the engagement ends.
Analytics data: only where you grant access, and only what is needed for the audit.
Public visibility signals: we look up how your brand appears across search engines, maps and business listings, review sites, knowledge graphs and company registries, and community sites. These providers receive limited data such as your brand name, domain and public URLs.
Competitor benchmarking: we also review publicly available competitor pages.
Anonymised research data
With your instruction under our agreement, we anonymise engagement data so it no longer identifies you or any individual, and use it to improve our methodology and benchmarks. See your Data Processing Agreement for details.
Who is responsible
You remain the data controller of all data you provide or grant access to. AireStream Ltd acts as your data processor, under a signed Data Processing Agreement. We are ICO-registered (registration ZC103113). For our own records (contracts, invoices, contact details) we act as an independent controller under our Privacy Policy.
Who touches your data (sub-processors)
We use a small set of vetted providers. Every AI provider below is used under commercial terms (API or business plans) that prohibit training on your data.
| Provider | What they do | Location |
|---|---|---|
| Hetzner Online GmbH | Hosts our audit engine and working data | Germany / Finland (EEA) |
| Anthropic | AI analysis (Claude API and Claude for Work) | USA |
| OpenAI | AI analysis (API) | USA |
| AI analysis (Gemini API, paid tier); search & visibility data (Places, PageSpeed Insights, Chrome UX Report, Search Console, Knowledge Graph); business email (Google Workspace) | USA | |
| Perplexity AI | AI search visibility testing (API) | USA |
| Microsoft | AI analysis (Azure OpenAI) and document storage / productivity (Microsoft 365) | USA / EU |
| Netlify | Hosting of client-facing report pages (benchmarks and assessments) | USA |
| Granola | AI meeting notes and transcription for client calls | USA |
| Slack | Internal team messaging about engagements | USA |
| DataForSEO | Search engine results (SERP) and SEO data for visibility analysis | EU (Estonia) |
| Ahrefs | SEO and backlink data for visibility analysis | Singapore |
| Trustpilot | Public review signals for the brand | USA / EU |
| Sentry | Error & performance monitoring (diagnostics; personal data redacted) | USA (EU region available) |
DataForSEO, Ahrefs and Trustpilot receive only limited identifiers such as your brand name, domain and public URLs.
Where a provider is located outside the UK or EEA, we rely on an appropriate UK transfer safeguard: UK adequacy for EEA providers, the UK Extension to the EU-US Data Privacy Framework, or the UK International Data Transfer Addendum (IDTA) / Standard Contractual Clauses (for example, Ahrefs in Singapore). Copies of the relevant safeguards are available on request.
We give clients at least 30 days' written notice before adding or replacing a sub-processor.
Within AireStream
Only AireStream's directors access client data on a need-to-know basis. We never enter your personal data or confidential information into consumer AI tools. Our internal AI Usage Policy restricts client data to the commercial AI services listed above.
How long we keep it
| Data | Retention |
|---|---|
| Search Console / Bing / analytics access | Revoked at the end of the engagement |
| Questionnaire responses, crawl data, analysis working data | 12 months after delivery in identifiable form (so a re-audit can measure progress), then anonymised. Anonymised data is no longer personal data and is retained for our research and benchmarking. |
| Data held by AI providers | Anthropic API 7 days (Claude for Work per Anthropic's commercial retention settings); OpenAI 30 days; Google Gemini API up to 55 days for abuse monitoring; Perplexity zero-retention |
| Audit deliverables and contractual records | 6 years (contract and tax requirements) |
| Error / diagnostic logs (Sentry) | Retained per provider settings; personal data redacted from diagnostics |
You can request earlier deletion of engagement data at any time, subject to legal retention requirements.

