← All postsHow-to

Shared drive structure: a folder tree people can guess

A shared drive structure fails when it mirrors the org chart. How to organise by work rather than by department, how deep to go, and what to do with the drive you already have.

How-toS

A shared drive structure is working when somebody who has never opened it can guess where a document lives on the first attempt. Most of them fail that test within a year, and the reason is almost always the same: the top level mirrors the organisation chart, so the moment teams are reorganised the structure describes a company that no longer exists.

Organise by work, not by department

  • Top level by enduring category — Clients, Projects, Operations, Finance, People — rather than by team name.
  • One level down by the thing itself: the client, the project, the year. These survive reorganisations because the work does.
  • Inside that, a small fixed set of folders repeated identically everywhere, so navigation becomes muscle memory.
  • No folder whose name is a person. People change roles, and their folder becomes an archaeological site.
  • No "Miscellaneous", "General" or "Stuff". Every drive has one, and it is where documents go to become unfindable.

How deep a shared drive structure should go

Three levels before you reach actual documents is the working limit — category, instance, section. Beyond that, people stop navigating and start searching, which is fine until they need to put something back, at which point the deep folder gets skipped and the file lands at the top. Depth is also where permissions get interesting: the deeper the tree, the more likely a subfolder has inherited access somebody no longer expects. A flatter structure with good names beats a precise taxonomy nobody maintains.

Date-prefix anything that recurs — 2026-Q1 rather than Q1 2026 — so alphabetical order is chronological order. It is a trivial convention that saves a surprising amount of scrolling, and it makes the stale material obvious at a glance.

Fixing a drive that already sprawled

  1. Do not reorganise everything. Create the new structure, move active work into it, and leave the old tree in place as an archive.
  2. Set a date after which the old tree is read-only, and tell people the date rather than asking them to migrate.
  3. Move whole folders rather than individual files, so history and permissions travel together.
  4. Check inherited permissions at the destination before moving anything sensitive.
  5. Write the structure down in one short page and put it at the top level of the drive itself.

What structure cannot do

It cannot stop duplicates, because people copy rather than link when they are unsure. It cannot make old material findable if nothing was ever named consistently. And it cannot substitute for deciding what gets kept — a beautifully organised drive holding eleven years of everything is still eleven years of everything. Structure is how people find things this quarter; retention is how the drive stays usable over a decade.

Ettex Storage supports the simple version well: folders with per-group permissions, a naming convention applied at the top level, and an access log so the quarterly review has something to read. Pair it with a file naming convention and most of the searching problem disappears without any taxonomy work at all.

Frequently asked

Should folders mirror the org chart?

No. Reorganisations are frequent and the work outlives the team structure. Organise by client, project or function instead.

How many top-level folders?

Five to eight. Fewer and they become meaningless buckets; more and nobody can hold the list in their head.

Is search enough on its own?

For finding, often. For filing, never — if people cannot guess where something goes, it ends up at the top level and the structure decays from there.

MI
Written by Maria I.

Part of the Ettex team — writing about product, engineering and the future of work.

More posts
Get the best of the Ettex blogProduct news, guides and tips — straight to your inbox, no spam.