Core concepts
Mycorr organizes everything in one hierarchy: an organization holds spaces, a space holds models, and a model holds tables. Learn these four levels and the ideas that connect them, and the rest of the product follows.
The four levels
Each level answers one question. Keeping the four questions apart is what lets a team share widely without losing track of who can see what.
| Level | Answers | Holds | Typical example |
|---|---|---|---|
| Organization | Who do we work with, and who pays? | Members, spaces, the plan and the bill | Your company, or your personal account |
| Space | Who can reach this work? | Models, and nested spaces | A team, a project or a client |
| Model | Where does the work happen? | Tables on a shared canvas | "Sales pipeline", "Budget 2027" |
| Table | What is the data? | Typed columns, rows and their history | Customers, Orders, Invoices |
Organization
An organization is the top of the hierarchy. It owns the data, holds the roster of members, and carries the plan and the bill: storage, egress and AI tool calls are all counted against the organization that owns the data, never against the person who clicked.
Every account comes with a personal organization that holds only you. Any other organization holds a team, with owners, admins and members. You can belong to several organizations and switch between them from the top of the sidebar.
Read more about organizations.
Space
A space divides an organization's work and decides who can reach it. You grant people a role on a space, as a manager, collaborator or viewer, and that role reaches every model inside it, including models created later. That is the point of spaces: one grant instead of a share for every model.
Spaces come in three kinds. Restricted and Sealed spaces sit at the top of the organization; a Sealed space also stops data from leaving it. Open spaces nest inside another space and inherit its access.
Model
A model is a workspace of related tables, laid out on a shared canvas where everyone working in it sees changes the moment they happen. It is the unit you open, arrange and work in, the way you would open a spreadsheet file or a database.
A model belongs to its space, and the space's managers own it. You can also share a single model with someone directly.
Table
A table holds your data in typed columns and rows, like a table in a database. Every table has one home model, where it is edited. Its columns can carry meaning as well as a type: a description, a unit of measurement, and whether the column is a key.
You look at a table through views. A table can have several views on the canvas at once, each with its own filters, sorting and visible columns, while the data stays in one place. See Views.
What connects them
The hierarchy says where things live. These ideas say how they relate, and they cross the boundaries between models, spaces and even organizations.
Two kinds of access
There are two ways to reach a model, and the difference matters when you plan how a team shares work:
- Through a space. A role on a space reaches every model in it. This is how a team works day to day.
- Through a direct share. You can share one model with a person, as a collaborator or a viewer, without giving them the rest of the space.
A single table can also be shared on its own. That never lets someone into your model: they import the table's release into a model of their own. See Sharing and permissions.
Live and released
Inside a model, everyone works on the latest version of its tables, and sees each other's changes as they happen. Outside it, people read a table's release: the version its editors chose to publish. That one rule is what lets you rework a table freely while someone else's report keeps reading the numbers you released.
The same rule holds for AI assistants: an assistant that can edit a table reads its latest version, and any other reads its release.
Where to go next
- Work through the Quickstart to see each level in practice.
- Learn the screens in Find your way around.
- Look up any term in the glossary.