Vendor ROI calculators tend to work the same way. Enter your staff count, accept some assumptions about hours saved per person per week, multiply by an hourly rate, and receive a number with a lot of digits. The number is not a lie exactly, but it was constructed backwards from a desired conclusion, and everyone in the room senses it.
The problem is not that measuring school software value is impossible. It is that it gets attempted with the wrong instrument, and the resulting figure collapses under the first serious question a governor asks.
The core difficulty is that hours saved are not money saved. If a system saves each of your forty teachers twenty minutes a week, that is roughly thirteen hours a week across the school and it produces no budget line whatsoever. You have not reduced staffing. What you have done is convert thirteen hours of administration into thirteen hours of something else — planning, marking, actually talking to a struggling student. That is genuinely valuable and it is not a cost saving, and describing it as one is exactly what makes the whole calculation feel dishonest.
A more defensible frame separates three kinds of return, and treats them differently.
The first is cash. Money that genuinely appears or stops leaving. Software licences retired because the new system replaces them. SMS costs reduced by routing through a cheaper channel. Reduced late tuition because reminders go out automatically and reliably — this one is real and measurable, and improvement in collection timing has an actual value you can calculate. Print and postage removed. This category is small, and it is the only part that belongs in a budget conversation without qualification.
The second is capacity. Time returned to people. Real, worth having, and not money. Present it honestly as what it is: the timetable took two weeks and now takes two days, so the deputy head got eight days back. Registers take three minutes instead of ten, so teaching starts earlier. Do not convert it to currency. The conversion is what makes the claim collapse, and the uncoverted version is more persuasive anyway because it is obviously true.
The third is risk reduction, which is the hardest to quantify and often the most important. A safeguarding record that cannot be lost. An audit trail that exists when a complaint arrives. Attendance patterns surfacing early enough to intervene before a child becomes persistently absent. Tenant isolation that makes a data breach structurally harder. You cannot put a credible number on any of these, and you should not try. State them as risk positions changed, and let the reader weigh them.
The single most valuable thing a school can do — and the one most consistently skipped — is take a baseline before the change. Before go-live, measure: how long does the timetable take, how many hours a week go to attendance follow-up, what proportion of fees are collected within thirty days, how long between an absence and the parent knowing. Five numbers, measured for two weeks. Without them, every subsequent claim is a guess, and the honest answer to "how do you know it helped" is that you do not.
It is worth being clear about what school software cannot claim. It does not improve learning outcomes directly. A management platform makes attendance visible, communication faster, and administrative load lighter; it does not teach anyone. There are plausible indirect chains — better attendance visibility leads to earlier intervention leads to fewer persistently absent students — and each link in that chain is a claim requiring evidence you probably do not have. Vendors who assert the endpoint without the chain are overreaching, and a school that repeats the claim to its governors inherits the problem.
Include the costs honestly too, which ROI calculations rarely do. Not just the licence: implementation time, training hours, the productivity dip during transition, the parallel running period, and the internal ownership the system needs ongoing. A calculation that counts every benefit and only the subscription fee is not an analysis.
And set a review date at purchase. Six months in, with the baseline in hand, ask what actually changed. Some things will have improved more than expected and some not at all, and knowing which is what makes the next purchase better. A school that has never gone back to check has no way to improve its own decision-making.
Skoolia surfaces operational measures — attendance patterns, collection timing, communication delivery, staff workload distribution — through analytics, which is what makes a before-and-after comparison possible at all. But the baseline has to be taken before the change, and only the school can do that.
A modest claim you can evidence will survive scrutiny in a governors’ meeting. A large one you cannot will not, and it damages the credibility of the next request.