Internal tooling
Gatekeeper
When four platforms become one problem.
- Client
- Sime Darby
- Role
- Product Design
- Key responsibilities
- Requirements gathering, Facilitation, Design
- Timeline
- 2 weeks
Disconnected platforms consolidated into one
Average response time for access requests
Admin task error rate (target: 5%)
Internal tools don't make headlines when they work. They make incidents when they don't.
The reality
Four separate platforms. Four separate user management systems. No single view of who had access to what.
Sime Darby is a large, multi-subsidiary business. And with scale comes sprawl. For the support and tech teams responsible for managing this, every day was a context-switching exercise.
Approving a new user meant logging into one system. Updating a role meant going somewhere else. Checking permissions across products? Manual, slow, and increasingly unreliable.
Admin errors were climbing, response times were dragging, and as Sime Darby continued adding products, the fragmentation was only going to get worse.
The first wrong move
Early temptation was to improve each platform individually.
Tighten up the UI here. Add better filtering there. Make each system less painful on its own terms. It felt pragmatic — work within existing infrastructure, avoid the complexity of a full consolidation.
But the more we dug in, the clearer it became: patching four broken systems just gives you four slightly less broken systems.
The root cause wasn't bad interfaces. It was the absence of a single source of truth.
The evidence
A whiteboarding session with the people who lived this problem every day.
Two tech leads and a support team member. Not a formal research study — a working session. Fast, direct, and grounded in real operational pain.
01
Too many platforms to manage
Managing users in separate systems created unnecessary complexity and increased the risk of oversight.
02
Difficult to scale
As more products were added, the fragmented setup made it harder to onboard and manage users efficiently.
03
Inefficient and high error rate
Without a single source of truth, admin errors became more frequent, potentially compromising platform security.
User access management sounds administrative. But get it wrong and you've got the wrong people with access to the wrong things. That's not an efficiency problem. That's a security problem.
The product needed to do one thing above everything else: give admins complete visibility and control, from a single place.
The solution
Built to reduce admin overhead and support scalability.
The goal wasn't to replicate what existed across four platforms. It was to design something that made those four platforms irrelevant.
Employee management. A clear employee list, instant access to profiles, role assignment in a single interaction.
Customer management. Review pending requests, assign roles, and action decisions without leaving the view.
Role management. Create and customise permission levels across products — consistent, documented, and auditable.
Scalability. Permission levels mapped to specific products and features, ready for every product Sime Darby adds next.
In the detail
Every summary view opens onto a real decision.
The screens above show the shape of the system. These are the moments admins actually spend their day in — declining a request, editing a profile, searching for the right role.
Declining a request. Rejecting a pending customer requires a documented reason — a small guardrail that keeps the audit trail honest.
Editing an employee. One click from the table into a full profile — job title, manager, and roles, editable in place.
Managing an active customer. Assigned companies and roles alongside removal — no separate system to open for either.
Assigning roles. Search-as-you-type role assignment, so admins never have to remember an exact role name.
What really happened
The error rate is the number that matters most. It's not a vanity metric.
It reflects whether the mental model landed — whether admins understood the system well enough to use it confidently and correctly. The support and tech teams described what was previously a fragmented, manual process as now centralised, faster, and more intuitive.
Disconnected platforms consolidated into one
Average response time for access requests
Admin task error rate (target: 5%)
Centralised, faster, more intuitive. It's so much easier to use and find what I need.
Key takeaways
01
Push for a quantified baseline
We knew errors were increasing, but "increasing" is hard to benchmark against. A documented starting point would have made the improvement far more concrete.
02
Spend more time on existing workarounds
The audit was valuable but compressed. A deeper look at how admins worked around the four systems would have surfaced nuances that only emerged post-launch.
Two weeks. One platform. Nobody's logging into four systems again.
Next case study
yourHD
When the biggest UX problem isn't the interface.