Your data stays yours
What happens to your data when we build an automation for you: where it runs, who can access it, how AI is used, and what we sign.
Where your automations run
We build in your own accounts: your Google Workspace or Microsoft 365, your spreadsheets, and your own n8n or Make account. If you prefer, we set up n8n self-hosted on your own server, so your data doesn't need to pass through our systems at all.
When we set up hosting for you, we choose an EU region by default.
Local AI for sensitive documents
If a workflow needs AI (for example to read invoices or summarise documents), the model can run on your own machine or server instead of an external AI provider. Sensitive documents then never leave your infrastructure.
Local AI is a custom add-on, quoted separately, because it depends on your hardware and on the documents involved.
Custom business systems →A data processing agreement (GDPR art. 28)
When we process personal data on your behalf (for example your clients' names, emails or documents), you are the controller and we act as your processor. We sign a data processing agreement under art. 28 GDPR before we access that data, and we act only on your documented instructions.
Least-privilege access, credentials in your accounts
We ask only for the access a job needs, and you stay in control of it.
- You invite us as a user or collaborator; we don't need your passwords.
- Credentials and API keys stay in your accounts, not in ours.
- At handover we remove our access and keep no copies of your data.
No training on your data
Your data is not used to train AI models. Where we use an external AI provider, we use it via API under business terms that exclude training on submitted data.
Sensitive workflows can avoid external AI entirely: many automations work with simple, transparent rules, and the rest can use local AI.
For accounting firms and other professionals
Accounting firms, consultants and other professionals are bound by professional secrecy towards their clients. We take that into account in the design: client documents stay in your accounts, reminders are sent from your own email, and we only see what the setup requires. Where you prefer, documents can be handled without any external AI.
What we don't claim
We're a small studio and we don't hold formal security certifications. Instead of badges, we commit to the concrete measures on this page and write them into the data processing agreement. As controller, you remain responsible for your own processing; we help you with the measures above.
Privacy policy (how we handle data on this website)Terms of sale
Questions about data and privacy
Does our data pass through your servers?
Not by default. Automations run in your own accounts (Google Workspace, Microsoft 365, your n8n or Make account) or on your own server with self-hosted n8n, so your data doesn't need to pass through our systems.
Where is it hosted?
In your accounts or on your server. When we set up hosting for you, we choose an EU region by default.
Do you sign a DPA?
Yes. When we process personal data on your behalf, we sign a data processing agreement under art. 28 GDPR before accessing that data.
Can we use AI without sending documents to an external AI provider?
Yes. Models can run locally on your machine or server, so sensitive documents never go to an external AI provider. This is a custom add-on, quoted separately. Many workflows don't need AI at all.
Is our data used to train AI models?
No. Where an external AI provider is used, it's via API under business terms that exclude training on submitted data.
Do you keep a copy of our data after the project?
No. At handover we remove our access and keep no copies of your data. Credentials stay in your accounts throughout.
We're an accounting firm bound by professional secrecy. Can this work for us?
Yes. Client documents stay in your accounts, emails go out from your own address, our access is limited to what the setup needs and is removed at handover, and documents can be processed without external AI.
Have specific data requirements?
Tell us what data is involved and where it must stay. We'll tell you whether it fits a package or needs a custom setup.