The bulletin
European by design: why we host our tooling in Europe
Where your network's management tooling runs matters more than where your network is. Here is how we think about it, what it costs us, and where the limits are.
When a company hands its office network to a provider, its data does not stay in the office. Monitoring data, device inventories, configuration backups, tickets, chat conversations — all of it flows to wherever the provider's tooling runs. For most providers, that means a stack of SaaS products, and most of those products are operated from the United States.
We made a different choice. Everything we use to manage a client's network — monitoring, documentation, IP address management, ticketing, support chat, backups — runs on open-source software that we host ourselves in data centres in the EU/EEA. Not because anything is wrong with American companies, but because of three practical consequences of where tooling runs.
Three reasons
Your GDPR position is simpler. Network data is more personal than it looks: IP addresses, device identifiers, usernames in logs. When it is processed in the EU by a processor with a short list of European sub-processors, the data processing agreement is a page, not a project. Transfers outside the EU/EEA need a legal mechanism and an assessment; the simplest way to keep that assessment short is to have nothing to assess.
You know where the data is. Self-hosted, open-source tooling can be audited, exported and, if needed, handed over. There is no vendor roadmap that removes a feature you rely on, no licence change that turns your documentation into a subscription, and no third-party support portal holding your configurations.
It keeps us honest about lock-in. Our tooling is ours to run, but the data in it is yours: accounts and licences are registered to your company, documentation is exported to you, configurations are yours. Hosting in Europe on software we control is what makes that promise cheap to keep.
What it costs us
More work. A self-hosted stack has to be patched, backed up, monitored and restored by us, not by a vendor's operations team. We think that is the right trade for a company whose entire job is running infrastructure well: if we cannot run our own tools, we should not be running your network.
Where the limits are
We manage the network, not your traffic: your office traffic does not pass through our systems, only management traffic does, over an encrypted tunnel per site. And where your own equipment is cloud-managed — Meraki, Aruba Central, UniFi's cloud — that vendor's cloud is part of the picture whether we like it or not. We do not hide it: it is listed as a sub-processor in your data processing agreement, and moving off it is an option we can cost for you, not a condition of working with us.
European by design is not a slogan about geography. It is a decision about who controls the tools that hold your data, and it is the one we made for ourselves before we asked anyone to trust us with theirs.