OperationsStorage

A company drive that turns documents into business records.

Use one company drive to store the evidence, extract the facts, and hand the result to the workflow that needs it.

A desktop-style file manager for the whole organization — folders, drag-and-drop, and shareable locations — that doubles as the front door of the books: send a receipt to intake and watch it get read, classified, and filed.

For teams whose files are not archives—they are the starting point for finance, HR, and operational work.

Storage · document intakeUploaded

dbs-statement-august.pdf

The file lands in an access-controlled organization folder. It is stored once, and everything downstream refers back to this object.

Storedorg folderComplete
Access checkedper objectComplete
OCRqueuedIn progress
ClassificationNot started
Drop a file in and follow what it becomes.
For Singapore operations teams

Keep the source file and the work it created together.

Storage combines a familiar company drive with controlled OCR and intake paths for Finance, HR, and operational records.

Org-scoped access
Keep every object under its organization path and check read or write access on the object itself.
Visible OCR state
Show pending, processing, filed, or failed status on the file so the team can see what happened without asking.
Canonical source
Run extraction against the stored object instead of creating a second processing copy that can drift.
Linked outcomes
Retain the connection when a statement becomes finance intake or a document becomes another platform record.
The actual screen

This is Storage.

Built from the product's own components and design tokens — the spacing, density, and status colours here are the ones you get after signing in.

theplatform.sharkflow.io/storage
Storage

orgs / finance / inbox

Files248
In processing3
Filed this month96
  • dbs-statement-august.pdf165 lines extracted · linked to a bank statement recordFiled
  • invoice-1042.pdfSent to finance intake · draft entry createdFiled
  • receipts-batch-aug.zipOCR running · 14 of 22 documents readProcessing
  • scan-0091.jpgUnreadable scan — needs a clearer copyFailed
Processing state lives on the file, not in a hidden job. Rendered with the product’s own components — sample records, not customer data.
A working file tree

The file does not stop at storage.

Use one company drive to store the evidence, extract the facts, and hand the result to the workflow that needs it.

Finance34 files
Employees18 files
Contracts12 files
Event briefs9 files

supplier-invoice-1042.pdf

Finance / Incoming

Processing

01 · Uploaded

Complete

02 · OCR

12 fields

03 · Classified

Supplier invoice

04 · Linked

Draft entry

Extracted text

Downstream record

Finance proposal #1042
Actions live beside the file

Drag in the files; decide what they should become.

The familiar folder model stays intact while extraction and workflow actions sit beside each document.

Feels like a desktop

Folder tree, breadcrumbs, drag-to-move, multi-select — every folder is a shareable link inside your org.

Uploads you can trust

Per-file progress chips show queued, uploading, and failed states — nothing disappears silently.

Documents that read themselves

Send a file to intake and OCR extracts the text; on finance folders, the bookkeeper classifies, renames, and files it.

Status on the file itself

Pills on each tile show intake progress — pending, processing, filed, or failed — with a legend that explains them.

Scoped to your organization

Every path lives under your org with access control on each object — not one shared bucket.

How it actually works

The mechanisms behind Storage.

Not features — the guarantees. Each one is enforced in the product rather than promised in a policy.

One canonical file
Extraction runs against the stored object. There is no second copy in a processing tool to fall out of sync with the one people can see.
Org-scoped paths
Every object lives under its organization’s own prefix, and read and write access is checked per object rather than per folder convention.
Visible pipeline state
OCR progress, extraction results, and bookkeeper status render on the file itself, so “did that go through?” is answerable by looking.
Evidence stays linked
When a document becomes a journal entry or a filing, the link is kept in both directions — the record knows its source and the file knows what it became.
Intelligence in Storage

Document pipeline.

Storage has no agent fronting it — it has a pipeline. Uploads are OCR-read and extracted on arrival, then made searchable by meaning rather than filename.

  • OCR and extraction run on upload, with a live pipeline you can watch
  • Semantic search retrieves by meaning, so a document is findable without its name
  • Extracted documents feed Finance intake directly, with no re-keying
Works with

Better together.

The modules share one organization record — connections are built in, not integrations to configure.

Questions from Singapore teams

Evaluating Storage? Start here.

Direct answers about local fit, controls, and how the module works with the rest of ThePlatform.

Is Storage a document management system or only file storage?
It is both a company file manager and an intake surface. Teams can organize files normally, then send supported documents into OCR, classification, and downstream workflows.
Can users see whether OCR succeeded?
Yes. Processing status appears on the file itself, including pending, processing, filed, and failed states.
How are documents separated between organizations?
Objects live under organization-scoped paths with per-object access checks. The public page uses sample records only and never exposes tenant files.
Can a source document remain linked to Finance?
Yes. When a supported file enters finance intake, the source and the resulting record retain links in both directions.

Let the company drive finish the paperwork it starts.

Put Storage to work inside the same platform as the rest of your operation.