For years, the most common sentence in public service delivery wasn't a complaint — it was a shrug. "Come back next week, we'll check." An application would disappear into a system of stamped registers, internal memos, and departmental silos, and the person who submitted it had no way to know if it was progressing, stuck, or lost entirely. The service existed. The visibility didn't.
This is the story of how that gap got closed — not through a massive overhaul, but through one deceptively simple idea: give every application a single, trackable identity from the moment it enters the system to the moment it's resolved.
The service existed. The visibility didn't.
The real cost of invisibility
On paper, the process worked. An application moved from intake, to verification, to approval, to issuance — four departments, four handoffs. In practice, each department kept its own records, in its own format, on its own timeline. A file could sit on someone's desk for eight days and nobody downstream would know to ask why, because nobody downstream could see it existed yet.
The visible symptom was long wait times. The invisible cause was that no single person — not the applicant, not the staff, not the supervisors — had a real-time view of where any given case actually stood. Delay wasn't a single bottleneck to fix. It was baked into the absence of shared visibility.
That diagnosis matches what institutions studying digital government keep finding. The World Bank's GovTech work on citizen-centered services treats service tracking as a transparency and accountability mechanism — not a nice-to-have notification layer — because tracking systems can identify who owns each step and who caused a delay. The OECD puts the same idea in user terms: people should not have to chase the state to learn what is happening to their own request.
Why adding more staff didn't help
The first instinct, understandably, was to throw people at the problem — more counters, more clerks, more approval staff. It helped a little, and only briefly. Backlogs shrank for a month and crept back within a quarter. Adding people to an invisible process just means more people working blind. The queue moves, but nobody can see the queue, so it fills right back up in a different place.
The actual fix had to attack the invisibility itself, not the headcount.
The shift: one reference number, one shared view
The redesign centered on a single change: every application, from the second it was submitted, was assigned one tracking number that followed it through every department, every handoff, every decision — visible to the applicant and to every internal team touching the case.
That one change did three things at once:
- It made delay visible in real time. A case sitting untouched for six days no longer hid inside someone's in-tray; it showed up on a dashboard as a case aging past its expected timeline, for anyone to see and act on.
- It made accountability specific instead of general. Instead of "the department is slow," a supervisor could see exactly which stage a case was stuck at and who owned that stage right now.
- It gave applicants a way to check status without calling anyone. A large share of front-desk traffic had never been about resolving problems — it was people simply trying to find out what was happening to their own application. Once that answer was available instantly, that traffic dropped on its own.
Visibility isn't a nice-to-have layered on top of a fix. Very often, visibility is the fix.
Automation handled the predictable, people handled the judgment
The tracking system alone would have just made the delay visible without shortening it. The second half of the fix was automating the parts of the workflow that were purely mechanical — routing a verified case to the next department automatically, sending a reminder when a case crossed its expected timeline, auto-populating repeated data instead of re-keying it at every stage.
None of that touched the parts of the process that genuinely needed a human decision — verifying a document, approving an exception, resolving a dispute. It simply removed the friction sitting around those decisions, so the people doing the judgment calls could spend their time on judgment, not on data entry and status-chasing.
That split — automate the mechanical, keep humans on judgment — is the same discipline McKinsey describes when governments harness automation at scale in application processing: the gains come from redesigned workflows, not from tools alone.
The result wasn't just speed
The most interesting outcome wasn't the drop in average processing time, though that mattered. It was the shift in tone. Complaints stopped being about "nobody tells me anything" and started being about specific, addressable issues — because for the first time, there was something specific to point to. A system that used to feel opaque and arbitrary started to feel, even to its harshest critics, like something that could be reasoned with.
The United Nations E-Government Survey 2024 measures maturity by whether people can actually use a service — find it, finish it, and understand what happens next. A portal that accepts an application and then goes silent fails that test even when the back office is working.
The lesson for anyone fixing a broken process
The instinct when a process is slow is to look for the slow step. Often, the real problem is that no one — not the people running it, not the people waiting on it — can actually see the process at all. Visibility isn't a nice-to-have layered on top of a fix. Very often, visibility is the fix, because it's what finally lets everyone else do their part.
Further reading
- World Bank — Service Upgrade: The GovTech Approach to Citizen-Centered Services
- World Bank — GovTech priority themes
- OECD — Good Practice Principles for Public Service Design and Delivery in the Digital Age
- UN DESA — E-Government Survey 2024
- McKinsey — How governments can harness automation at scale
- NeGD — National e-Governance Division
