Most institutions did not choose fragmentation. They accumulated it.
A learning platform was adopted for one purpose. A student records system was already in place. Assessment was handled somewhere else because the existing tool could not do it. Compliance reporting was built in a spreadsheet by someone who understood the requirement, and it has worked ever since, which is both fortunate and precarious.
Every one of those decisions was defensible at the time. Together they produce an operating environment where no system holds the whole picture and the connective work is done by people.
The hidden cost of the gaps
The expense of fragmentation is rarely visible as a line item, because it is distributed across the people who absorb it.
It appears as the same information entered more than once, in more than one place, with the inconsistencies that follow. It appears as reconciliation — the recurring work of establishing which of two conflicting records is correct. It appears as reporting that takes days because the underlying data has to be assembled before it can be analysed. It appears as decisions delayed because nobody is confident enough in the numbers to act on them quickly.
It also appears as institutional fragility. When the integration between systems lives in one person’s understanding of how the parts fit together, the institution is exposed to that person’s absence.
Ask an administrator how much of their week is spent moving information between systems that should already agree. The answer is usually substantial, and it is almost never in anyone’s plan.
Why adding another tool rarely helps
The instinctive response to a capability gap is to acquire a tool that fills it. Sometimes that is correct. Often it makes the underlying problem worse.
A new point solution solves a specific problem well and introduces a new boundary: another login, another data set, another export, another place where the institution’s version of the truth can diverge. The capability improves; the fragmentation deepens.
This is why I am wary of evaluating education technology purely on features. A feature comparison measures what each system does in isolation. It does not measure what the institution has to do to hold them together — which is where most of the cost actually sits.
The more useful question is not “what does this add?” but “what does this remove?” A platform that adds a capability and removes three reconciliation steps is a different proposition from one that adds a capability and creates a fourth.
What an operating system means here
When I describe the goal as an operating system for education, I am not making a claim about scale. I am describing a role.
An operating system is not the most visible software on a machine. It is the layer that manages shared resources, enforces consistent rules and lets everything above it assume a coherent environment. Its value is that other things do not have to solve the same problems repeatedly.
Applied to an institution, the same idea means a layer where the shared facts live once: who the learners are, what they are enrolled in, what they have been assessed on, what the outcomes were, what the institution is obliged to report, and who is permitted to act on any of it. Learning, administration, assessment, compliance and student management stop being separate systems that must be kept in agreement, and become views onto the same underlying record.
The change that produces is not primarily about convenience. It is about what becomes possible. An institution with one coherent view of itself can answer questions it previously could only estimate — where learners are actually dropping out, which programmes complete, whether an accreditation cycle will be survivable before it begins, where staff effort is really going.
That is the difference between a system that stores information and a system that supports decisions.
Integration is not the same as consolidation
I want to be careful here, because “one integrated platform” can be heard as “one vendor for everything”, and that is not the argument.
Institutions have existing systems that work, contractual commitments, and specialist requirements that a general platform will not serve better. A strategy that requires replacing all of it is not a strategy; it is a demand.
What matters is where the authoritative record sits and how cleanly the boundaries are drawn. A specialist tool connected to a coherent system of record is a very different situation from four systems each holding a partial version of the truth. Integration in the sense that matters is about coherence, not about ownership.
The practical test is simple: when two systems disagree, is it clear which one is right? If the answer requires a discussion, the institution does not have an integrated environment regardless of how many integrations exist.
Building it is a different discipline
There is a reason more products do not attempt this, and it is not lack of ambition.
Point solutions can be opinionated. They serve a narrow problem and can assume a great deal about how it is approached. A system of record cannot. It has to accommodate institutions that are genuinely different — different qualification structures, different approval chains, different regulators, different operating models — without dissolving into a bespoke build for each one.
That constraint shapes everything. It pushes towards a common architecture with configurable process rather than fixed workflow. It demands that the data model be right early, because it is the hardest thing to change later. It requires resisting the pull towards features that demonstrate well but fragment the model.
It is slower. It is also the only version of this that scales, because a platform that requires a bespoke implementation per institution is a services business wearing a product’s clothing.
Where the value accumulates
The strategic argument for building at this layer is that its value compounds while a point solution’s does not.
Interfaces are replaced regularly. Fashions in learning technology change. But the system that holds an institution’s records, enforces its processes and answers for its compliance obligations becomes more valuable the longer it operates, because the record it holds becomes more complete and more relied upon.
That is a durable position, and it is earned rather than claimed. It requires being correct, being available, and being trusted with things that matter — over years, not quarters.
The direction of travel
I do not think the next wave of education technology will be defined by better learning content. Content has improved enormously and will keep improving.
I think it will be defined by whether institutions can finally operate from one coherent view of themselves — and whether the platforms serving them reduce fragmentation or quietly add to it.
That is the standard I would want applied to anything built for this sector, including our own work. Not: what does this system do? But: how much of the institution’s connective work does it remove?