Why recent media coverage should prompt reflection across the sector
Recent reporting on assessment failures at the University of Glasgow has understandably drawn significant attention across the assessment sector and beyond.
Much of the coverage to date has focused on outcomes and accountability. That is natural. But it is also only part of the story.
For any organisation – whatever its size – delivering assessments today, the more uncomfortable and ultimately more useful question is not who is at fault, but how such failures can occur without being detected earlier.
This article is not about pointing fingers, nor about revisiting the personal circumstances at the heart of the case. Instead, it is an invitation for all of us to pause and reflect on the structural and operational risks that exist, often quietly, within our assessment processes.
Assessment failures are rarely single points of error
One of the risks of bite-sized media reporting is the impression that assessment failures result from a single mistake, a single individual, or a single broken link in the process.
As most of us know, in reality that is rarely the case.
When things go wrong, it is usually the result of a series of small broken links, none of which — in isolation — appear likely to cause the chain to fail. This might include:
- Manual workarounds that have become standard practice
- Too many unchecked hand-offs between teams
- Data pulled from multiple systems into spreadsheets
- Too few staff during peak periods
- Inconsistent application of rules under pressure
- Inexperienced staff operating in stressful environments
- Complacency around the verification of outcomes
- Evidence that exists, but cannot easily be reconstructed
Individually, each of these links feels tolerable. Collectively, they create conditions in which errors can persist unnoticed, sometimes for long periods. And eventually, the chain snaps.
Quiet risk is easy to live with — until it isn’t
I suspect that almost everyone reading this article is confident that their processes “work”. And in most exam cycles, they do. The chain has held, and outcomes have been delivered.
There are, however, two points worth considering.
First, any chain with a weakness will break eventually. Not always predictably, and rarely conveniently.
Second, most risk does not sit in the day-to-day operation — the 90% of cases that flow through exactly as intended. It hides in the remaining 10%: the exceptions, and within periods of peak assessment stress.
These are precisely the moments when:
- The people who know how things really work matter most
- Decisions and evidence need to be clear, recorded, and defensible
When assessment systems rely heavily on what sits in someone’s head, or on a series of manual reconciliations with limited oversight, confidence is often higher than resilience.
Reporting often underestimates operational complexity
From the outside, assessment failures are often viewed from a simple perspective. Candidates, clients, the media — and sometimes even legislators — may assume that modern assessment providers operate fully integrated, automated, and transparent systems.
Those working behind the scenes know this is not always the case.
Assessment processes are frequently layered over decades:
- Legacy systems coexist with newer tools
- Manual workarounds intended as temporary fixes become the norm
- Processes evolve incrementally, rather than by design
None of this implies poor intent or lack of care. It reflects the reality of complex organisations operating under constraint.
But it does mean that risk can become embedded, and, over time, invisible.
Time for reflection, not reaction
For leaders across assessment organisations, the most constructive response to recent events is not reassurance or defensiveness, but active curiosity.
Now is the right time to ask some uncomfortable but valuable questions internally:
- Which processes depend on specific individuals knowing “how it’s done”?
- What happens when those individuals are unavailable during peak periods?
- How many manual workarounds and hand-offs currently exist in our delivery model?
- How quickly could we reconstruct the full decision trail for a contested result?
- Where does evidence actually live — not where we assume it lives?
- How do we test our assessment processes under stress, not just at steady state?
These are not questions prompted by media coverage. They are questions of leadership.
Assessment is infrastructure — and infrastructure fails silently
Where assessment matters, it supports progression, professional standing, and trust. Not just in institutions, but in the professional communities they serve.
When it works, assessment is invisible. When it fails, the consequences can be profound.
The lesson from recent media coverage is not that assessment systems are uniquely fragile, but that their fragility is often underestimated — both internally and externally.
Improving resilience does not require radical transformation in every case. Often, it begins with simple steps:
- Clarifying decision ownership
- Reducing manual workarounds and reconciliation
- Making evidence retrieval the rule, not the exception
- Designing processes that can support peak periods of activity, not average demand
A shared responsibility across the sector
We should not leave this moment assuming that risk is confined to one institution or one part of the sector. That is unlikely to be the case.
A more responsible response is collective learning, recognising that assessment systems carry real human, professional, and reputational weight. Managing that burden requires ongoing attention to how work is actually carried out, not just how we assume it operates, or how we are told it works.
If recent events encourage organisations to look more closely at their own assessment processes, particularly where risk is quiet and taken for granted, then something constructive may yet emerge from a very difficult period.
