Three weeks into her maternity leave, the registrar’s office at one private university discovered that nobody else could run the postgraduate fees calculation. The spreadsheet she had built over four years contained eleven sheets, cross-referencing scholarship eligibility against programme codes and citizenship status, with a column of formulas nobody else had ever needed to understand. Her colleagues called her the following week to ask how to update it for the new semester. She apologised and walked them through it from home, which took ninety minutes and required her to log into the institution’s VPN.
She wasn’t paid for that time. The institution didn’t really register that it had asked her to do this. But for the next eight months, until she returned, the postgraduate fees process ran on phone calls and shared screens, because the official campus management system did not, in fact, calculate postgraduate fees correctly. The spreadsheet was where the actual calculation lived. Nobody had decided this. It had just happened.
Why universities run on spreadsheets
Every university has these. Some have hundreds. Most don’t know how many.
The spreadsheets emerge wherever the official system doesn’t quite do what an operational team needs. The campus management platform was implemented six years ago and the fee structure has shifted twice since then, so somebody in finance builds a spreadsheet to handle the new categories. The scholarship eligibility rules are too nuanced for the configured workflow, so somebody in the registrar’s office builds a spreadsheet to track them properly. The graduation audit needs to cross-reference programme structures against student records across three intakes, so somebody in academic services builds a spreadsheet to do it.
Nobody approves these spreadsheets. They’re built by people who needed to do their jobs and discovered the system wouldn’t let them. They’re shared by email, stored on personal drives, and quietly become the operational layer the institution actually runs on. The official system continues to be the system of record. The spreadsheets become the system of practice.
This is reasonable behaviour. The staff who build them are doing what staff have always done, which is solving the problem in front of them. The cost only becomes visible when somebody asks how the institution actually operates, or when the person who built the spreadsheet stops being available.
The risks that don’t show up in a procurement document
The risks of spreadsheets in higher education are a specific class of operational and governance exposure. They are rarely accounted for inside the institution, and they almost never appear in the original case for buying the campus management system.
Most of them come back to one staff member. A spreadsheet built by one person usually depends on that person to maintain it, interpret it, and explain it. When they leave, retire, or simply move teams, the institution loses operational capacity it didn’t know it had until it was gone.
Even when that staff member stays, the data inside the spreadsheet rarely stays trustworthy. Spreadsheets are easy to edit and difficult to validate. A finance team running fee calculations on a shared spreadsheet has no straightforward way to know whether a formula has been changed, whether a row has been deleted by accident, or whether the version they’re working on is the latest one. Errors surface as anomalies in reporting, often months after the fact.
The same spreadsheet is usually also a data-security problem the institution hasn’t classified as one. The file contains the same sensitive student information the official system does, but with none of the access controls, audit trails, or backup arrangements. A scholarship eligibility spreadsheet emailed between finance and the registrar’s office is a personal-data exposure waiting to happen, and most institutions cannot tell you how many such spreadsheets exist on their staff laptops at any given moment.
And when the regulator or external auditor asks how a particular institutional process actually works, the answer that traces back to a file on someone’s machine is not a satisfactory answer. The institution may be doing the right thing operationally. It cannot demonstrate that it is.
The institutional cost
These risks compound. A university that runs significant operational processes on spreadsheets accumulates a quiet operational fragility that doesn’t appear in any dashboard.
Decisions get slower because every cross-departmental question requires somebody to reconcile multiple spreadsheets manually. Reports take longer to produce because the underlying data lives in places the reporting tools can’t reach. Audits take longer to prepare because the institution has to first work out which spreadsheet is the authoritative version of each process.
And governance drifts. The pattern, from a different angle, is the same one that appears when governance lives on paper: enforcement depends on whoever is on duty. A spreadsheet is just a smaller, more portable version of that problem. It encodes a process. Nobody validates what it does. Nobody enforces what it does. Over time it drifts from whatever was originally intended, and the institution doesn’t notice because the spreadsheet still runs.
The cumulative cost is rarely calculated, because no single department bears it. It is paid in administrator hours, in maternity-leave phone calls, in audit-preparation overtime, and in the occasional dramatic incident: the wrong fee charged at scale, the data leak through an email forward, the year-end reconciliation that takes three weeks instead of three days.
Why “just don’t use spreadsheets” is the wrong response
The instinct, once an institution recognises this pattern, is to ban operational spreadsheets. This almost never works.
Spreadsheets exist for a reason. They fill gaps in the official system. Banning them doesn’t close the gaps. It just forces staff to work around those gaps less visibly, often by maintaining the same data in email threads, shared documents, or even private notebooks. The institutional risk doesn’t decrease. It just relocates.
The right response is to treat the existence of a workaround spreadsheet as a diagnostic signal. Every spreadsheet that an operational team relies on is the system telling the institution where it has failed to meet operational reality. The work is not to eliminate the spreadsheet. It is to absorb its function into the platform, so that the platform finally does what the staff had been doing manually.
This is harder than it sounds, because the spreadsheet often encodes institutional logic that nobody has ever written down. The work of absorbing it is the work of writing that logic down, validating it with the people who built the spreadsheet, and configuring it into the system as a proper workflow. The spreadsheet then becomes obsolete because the system has taken over what it was doing.
Anything else just moves the workaround somewhere harder to see.
Where this leads
The institutions that take this seriously start by counting. They ask each operational team to identify the spreadsheets they actually depend on, and they triage from there. The ones encoding routine calculations get absorbed first. The ones encoding institutional policy get prioritised, because they represent the same kind of governance theatre: documented in one place, enforced in another, traceable in neither.
The institutions that do this consistently report a different pattern. Their staff still build spreadsheets, but the spreadsheets are temporary, used for analysis, for ad-hoc reporting, for individual investigation. They are not the operational backbone of any institutional process. When a member of staff goes on leave, the work continues. When the regulator asks how a process operates, the answer points to a configured workflow inside the platform, not to a file on somebody’s laptop.
The registrar from the opening of this piece eventually returned from maternity leave. The fee calculation was no longer her spreadsheet by then. The institution had spent the second half of her leave absorbing it into the campus management system, configuring the postgraduate fee logic into a proper workflow, and validating the result against her records. When she came back, she didn’t have to maintain the spreadsheet anymore. The institution had finally caught up to what she had been doing alone for four years.
