Why Office Unbound
Why build another
productivity platform?
Because almost every business now runs its email, documents, files and meetings on one of two subscription platforms, on their terms — and for a meaningful share of organizations, a different arrangement would serve them better.
Project status Office Unbound is in Phase 1, architecture and evaluation. This page explains the reasoning behind the project; nothing it describes is available yet.
The problem it addresses
Not that Microsoft 365 and Google Workspace are bad. That there is so little choice of architecture.
Twenty years ago a business that wanted email, shared files and calendars ran a server, or paid someone to run one. Today it buys a subscription. That shift brought real benefits — no hardware, no patching, features that improve every month — and it also brought a single model for everyone: the vendor’s infrastructure, the vendor’s release schedule, the vendor’s licensing terms, and a service that stops when the subscription does.
For many organizations that trade is the right one and will stay the right one. For others it creates problems they cannot solve from inside the subscription. An MSP can resell Microsoft 365 but cannot shape it, so its service looks like every other MSP’s. A business that needs its data in a particular place, or on a network it controls, chooses from the options the vendor offers. A security team asked to approve the platform has to rely on the vendor’s assurances about code it cannot read. And when per-user prices or bundles change, the customer’s only real choice is to accept or to undertake a migration.
The open-source building blocks for an alternative already exist, and many are mature: mail servers, office editors, file platforms, messaging systems, identity providers. What does not exist is a platform that brings them together. Each one has its own login, its own administration console, its own update cycle and its own idea of who a user is. Assembling them into something a business can use every day is a significant project that each organization or MSP currently has to undertake by itself.
Office Unbound is being built to be that platform — one identity, one interface, one administration layer, multi-tenant provisioning and a migration path — so that running your own productivity platform becomes a choice an ordinary business or MSP can realistically make.
Five principles, made concrete
Principles are easy to state and easy to ignore. Each of these is tied to specific design decisions, and each has a cost that is stated alongside it.
- 01Ownership
-
Your data sits on storage you or your chosen provider control, with exports in formats you can read without Office Unbound: mail as standard message files, files as ordinary files, documents in Office and OpenDocument formats, calendars and contacts in iCalendar and vCard. Export is designed in from the start rather than added as a compliance feature.
The cost: owning data means owning its backups. Nobody else is keeping a copy for you. Ownership and independence
- 02Independence
-
No single company — including the Office Unbound project — should be able to switch your platform off, change its price or force a migration. Open-source licensing means the software keeps working if the project changes direction; open standards mean data and integrations are not tied to one implementation; a separate operator layer means you can change MSP without changing platform.
The cost: independence is not free of dependencies. You depend on upstream projects, your operator and your own staff instead of one vendor.
- 03Control
-
The organization operating the platform decides where it runs, how its network is segmented, which workloads are enabled, which policies apply and when updates reach production. MSPs can set those per customer. Control extends to the update schedule: releases are staged and approved, not pushed.
The cost: control over updates is also responsibility for applying security fixes promptly. Security and transparency
- 04Transparency
-
Source code, dependencies, build provenance, security advisories and the roadmap are intended to be public. So is the project’s status: every capability on this site carries a label, and nothing is described as available until it has been verified in a release. Today nothing has.
The cost: transparency makes weaknesses visible as well as strengths. That is the point, but it is uncomfortable.
- 05Choice
-
Customer-hosted or MSP-hosted. Which workloads to run. Which office editor or mail server, where the platform supports alternatives. Whether to keep some services on Microsoft 365 or Google Workspace during a long coexistence. The platform is modular so that adopting it does not have to be all-or-nothing.
The cost: every choice is a decision someone has to make and support. Sensible defaults are part of the design.
The real trade-offs
Running your own platform — or having an MSP run it — is not simply a better version of a cloud subscription. It moves work and risk to different places. This is how they compare in general terms.
| Consideration | Cloud subscription | Self‑operated or MSP‑operated |
|---|---|---|
| Who runs it | The vendor, with very large operations and security teams. | Your IT team or your MSP. Quality depends on them. |
| Security operations | Strong, mature and largely invisible: threat detection, patching and incident response at scale. | Your operator’s responsibility, supported by the platform’s controls and advisories. Inspectable, but only as good as its operation. |
| Compliance evidence | The vendor holds certifications and publishes audit reports customers can rely on. | No vendor certification to inherit. The organization establishes its own compliance, using the platform’s controls. |
| Data export | Good tools exist — eDiscovery and admin exports, Takeout — in the vendor’s formats. | Data is already on your storage, in open or standard formats. |
| Cost structure | Predictable per-user fees, set and changed by the vendor. | Infrastructure, storage, backup, labor and optional services. Not necessarily lower; differently controlled. |
| Updates | Continuous, on the vendor’s schedule. | Staged and approved by the operator, who must apply security fixes promptly. |
| Outages | Rare, but when they happen the customer can only wait. | Possibly more frequent with weaker operations, but within the operator’s power to diagnose and fix. |
| Data location and network | Chosen from the vendor’s regions and options. | Wherever the operator puts it, on networks it designs. |
| Feature breadth | Very wide, with large third-party ecosystems. | Narrower. Office Unbound targets the core workloads, not every feature. |
| Time to start | Minutes. | Days to weeks, plus migration. |
Neither column is better in every row. The question is which rows matter most to your organization. The tables scroll sideways on narrow screens.
Microsoft and Google are not the problem this project is solving. Lack of choice is.
It would be easy to market an alternative by caricaturing the incumbents. It would also be inaccurate. Microsoft 365 and Google Workspace have strong identity protection, mature threat detection, extensive compliance tooling — retention, eDiscovery, data-loss prevention — and real export capabilities. Their security teams are larger than most companies.
Office Unbound’s case rests on different ground: who operates the platform, who can inspect it, who decides when it changes and how easily a customer can leave. Where an organization values those more than the breadth and convenience of a subscription, an alternative should exist. Where it does not, the subscription is the better choice.
Who it is not for
Saying this plainly saves everyone time.
- Organizations with no one to operate itIf neither in-house IT nor an MSP can run servers, monitor services and handle backups, a cloud subscription is the safer choice.
- Anyone who needs it working todayNothing is available. Phase 1 is architecture and evaluation, and no release dates are published.
- Deep Power Platform or Apps Script usersAutomations, custom apps and add-ins built on Microsoft or Google platforms do not move and will need rebuilding.
- Heavy users of advanced Office featuresComplex Excel models, VBA macros and intricate Word templates may not behave identically in open-source editors.
- Organizations relying on inherited certificationsIf compliance depends on the vendor’s SOC 2 or ISO reports, self-operation means building that evidence yourself.
- Anyone choosing purely on priceOffice Unbound may cost less for some organizations and more for others. It promises neither.
Who it is for
The organizations Office Unbound is being designed around have something in common: the architectural questions matter to them as much as the feature list.
- MSPs that want to build and own a productivity service rather than resell one.
- Businesses with an MSP or IT team they trust, and a reason to want more control.
- Organizations with data-location, network-isolation or sovereignty requirements.
- IT and security teams that want to inspect and test what they run.
- Developers and integrators who want open APIs and open standards underneath.
Where it stands
The project is in its first roadmap phase Current phase: selecting the open-source components, reviewing their licenses, designing the security architecture and integration layer, and designing how migrations will be assessed. Core services, collaboration, the client experience, the migration platform, MSP tooling and hardening follow in that order.
No dates are published, because none could be supported yet. The roadmap will change as evaluation produces evidence, and the project updates page will record why.
Does the trade make sense for you?
Tell us what you run today and what would have to be true for you to change. Early interest shapes what is built and tested first.