EDMS: Document Management System
A system for how organisations store, organise, track and retrieve documents, built for the way offices actually work.
- Role
- Lead Product Designer, GX Informatics
- Client
- GX Informatics
- Industry
- Enterprise · Information management
- Year
- 2025
Project overview
Most corporate offices in Nigeria still run on physical cabinets and shared drives. I designed EDMS, a system for how organisations store, organise, track and retrieve documents, built for the way offices actually work.
Context & problem
“They weren't mismanaging documents. They just had no system.”
Nobody wakes up and decides to mismanage documents. Cabinets become overcrowded. Files get shared over WhatsApp because it's faster. A folder goes missing. An office fire destroys years of records overnight.
That was the reality for the offices I was designing for. Poor document handling wasn't just an operational issue; it was a risk to continuity, accountability and trust.
Who it serves
Auditors were a revealing group: they weren't interested in browsing. They wanted to search, find a file and get out.
Publisher
Creates and uploads documents.
Librarian
Organises and retrieves them.
Admin
Controls access and settings across the system.
Project goals
- Make migrating existing paper records realistic.
- Mirror the mental model people already use.
- Give the right people the right access, and no more.
- Make finding a document fast.
Research & discovery
Speaking with stakeholders, I learned people weren't going to abandon a mental model they'd relied on for years. The goal was never to reinvent how they thought about documents; it was to make switching feel easy.
Those sessions showed that not everyone uses a system the same way, so search became a first-class feature.
And the thing that almost killed adoption had nothing to do with the interface. It was the scanning process: scan, extract text, upload. Three separate steps nobody wanted to do fifty times a day.
Constraints
Thousands of documents already lived in physical cabinets. If we built a powerful digital system and ignored migration, adoption would fail, and staff would be stuck between paper records and digital workflows.
Access was a balancing act: too much and we recreate the original chaos; too little and the system quietly fails through lack of use.
Product decisions
A two-phase approach
First make migration easy, then structure everything around what people already know.
Batch scanning + OCR in one action
Scanning, text extraction and upload collapsed into a single step, which removed the biggest reason not to use the system.
Department → Cabinet → Folder
The information architecture mirrors the physical office instead of asking people to relearn everything.
Role-based access, mapped with stakeholders
Together we mapped who needs to see what, who needs to act on what, and where the boundaries sit.
Information architecture & workflows
- Step 01Department
- Step 02Cabinet
- Step 03Folder
- Step 04Document
Exploration & iterations
The choices that decided whether people would actually switch.
Migration first, not features first
I pushed for a two-phase approach, with batch scanning and OCR added even though they weren't in the original scope.
WhyWithout an easy way to move thousands of paper files, adoption would have failed quietly.
Familiar structure over a new one
Department → Cabinet → Folder, mirroring the physical office.
WhySwitching should feel easy, not like relearning everything.
Search as a first-class feature
Auditors didn't want to browse. They wanted to search, find a file and leave.
WhyNot everyone uses a system the same way.
One action instead of three
Scan, extract text and upload collapsed into a single step.
WhyNobody wanted to repeat three steps fifty times a day.
Access: not too much, not too little
Mapped with stakeholders: who needs to see what, and who needs to act on what.
WhyToo much access recreates the chaos; too little and the system goes unused.
Final solution
Collaboration & implementation
Working with stakeholders, not just for them.
Stakeholder sessions shaped both the folder structure and the access model. We mapped permissions together rather than designing them in isolation.
I also made the case for expanding scope to include batch scanning and OCR, and designed role-based access and batch PDF export, turning a repetitive manual task into a single action.
After launch, the client's requests, starting with folder access by role and password, fed directly into the V2 roadmap.
Results & evidence
The system went live, and stakeholders described it as seamless.
Soon after, they asked to configure folder and file access by role and password. That kind of request only comes when people have stopped thinking about the old way and started thinking inside the new system.
Lessons & reflections
Next: granular folder-level permissions, AI-assisted classification built on the OCR foundation, and an audit-trail dashboard that makes a compliance review something you can do in under five minutes.
The biggest barrier may not be on screen
The interface wasn't what threatened adoption. The scanning process was.
Familiar beats clever
Mirroring the physical office made switching feel easy.