Everything works on presentation day: training is available, a representative from the supplier is in the room, and time has been reserved. The difficult test comes six months later, when a teacher leaves, a licence expires or a family cannot sign in. That is when a school discovers whether innovation has become organisational capacity or still rests on the enthusiasm of two people. Maintaining educational technology takes much more than renewing it: the school must know what it is for, who sustains it and when it should be abandoned.
The mature question is not whether the school looks modern, but whether it can provide a stable, fair and understandable experience. A useful platform may remove repetitive work, improve accessibility or make materials easier to reach. It can also add passwords, notifications, technical dependence and invisible labour. Rigorous evaluation is not opposition to technology. It means treating it like other educational infrastructure, with objectives, a budget, accountable owners, maintenance and a right to review.
A pilot does not prove that a system is sustainable#
A pilot answers a limited question: can this solution produce a benefit under particular conditions? It does not prove that it will work across every year group, that staff will have time, that vulnerable pupils can participate or that costs will remain reasonable. Trials often concentrate support, new devices and motivated people. Wider deployment brings more diverse classes, simultaneous incidents and weeks when nobody can pause to resolve an update.
A pilot answers a limited question: can this solution produce a benefit under particular conditions?
Before expansion, record the baseline. What precise problem existed? How much time did it consume? Who was excluded? Which educational outcome was expected to improve? Without this picture, any novelty can look like progress because it creates activity. A useful indicator is not the number of logins but an outcome connected to purpose: fewer lost assignments, more accessible materials, faster feedback or greater independence. Unintended effects, such as night-time family support or duplicate record keeping, matter too.
The comparison must include a reasonable alternative, rather than setting the new tool against a deliberately poor practice. A digital diary should be compared with a well-managed diary; automated feedback with a feasible teacher-led routine. The task is to discover which option best addresses the problem with available resources. Novelty earns no bonus, and neither does familiarity.
UNESCO examines educational technology through relevance, equity, evidence and sustainability in its Global Education Monitoring Report on technology in education. This supports a simple rule: a tool deserves to continue only when its educational benefit is demonstrable, it reaches the learners it should reach, and the organisation can sustain it without consuming other essential resources.
Turn personal knowledge into the school's memory#
The clearest warning of fragility is, “Only Marta knows how that works.” The task may be creating accounts, recovering work, configuring permissions or adapting materials. If the procedure lives in one person's memory, it is not institutional capacity. Frequent tasks need short instructions, an update date and an owner. Someone who did not join the pilot should test the documentation; if they cannot follow it, the process still depends on tacit knowledge.
A minimum inventory records the tool, purpose, affected classes, educational owner, technical owner, supplier, types of data, cost, renewal date and alternative. This is not decorative bureaucracy. It supports action when leadership changes, a substitute arrives or an incident happens during a holiday. It also exposes duplication: three messaging apps, two repositories for the same files, or paid licences that almost nobody uses.
Every critical function needs at least two trained people. They need not become professional administrators, but they should understand the user journey, resolve common incidents and know when to escalate. Allocating working time is part of the design. If maintenance happens during breaks and at night, the apparent saving is funded by invisible labour and will end when that person becomes exhausted or leaves.
The European SELFIE platform helps schools reflect on technology use with input from leaders, teachers and pupils. Its value is not a decorative score but the comparison of perspectives: leaders may see a stable implementation while staff and students report confusing access or weak support. Repeating the reflection helps determine whether agreed improvements reached the classroom.
Budget for the full cost, including time#
The licence price is only one component. Devices, connectivity, cases, chargers, storage, backups, support, training, cover staff, accessibility, migration and coordination time all cost something. There is also opportunity cost: an hour spent rebuilding accounts cannot be spent on tutoring or lesson preparation. The calculation should cover several years, because a cheap introductory offer may become expensive once a school depends on the format and moving its data is difficult.
Assign each cost to the people who actually bear it. If a school application requires families to supply a phone, data and daily assistance, part of the budget has moved outside the school. If a teacher privately converts inaccessible documents, the cost appears as personal overload. Making this distribution visible supports fair decisions and directs funded help to where it is needed.
Renewal should not be automatic. Months before the contract expires, the educational owner presents use, benefits, incidents, costs and alternatives. People who stopped using the system or took the alternative route must be heard; studying active users alone creates a misleading picture. The decision may be to retain, limit, renegotiate, replace or remove. Continuing because migration feels frightening is dependence, not evaluation.
Sustainability includes material impact. Extending equipment life, repairing devices, sharing suitable hardware and avoiding unnecessarily high requirements reduce expense and waste. Software that forces replacement of functioning computers needs to justify a substantial benefit. Responsible procurement asks about updates, energy, repairability and compatibility before classrooms are filled with equipment that will soon lose support.
Privacy, accessibility and continuity are not extras#
Data protection requires review throughout a tool's life, not only on the day of purchase. Functions, permissions and suppliers change. The Spanish Data Protection Agency's guide for educational centres provides criteria for handling data in schools. The institution must be able to explain what information is collected, for what purpose, who can access it, how long it is retained, and how it can be deleted or exported.
An update that activates profiling, additional analytics or artificial-intelligence functions deserves a fresh assessment. The convenience of pressing “accept” is not a decision. A personal account should not be demanded when an educational account can be provided. When a family raises a concern, it needs an understandable answer and an accountable person, not a chain of links to legal conditions detached from the activity.
Accessibility is tested by performing real tasks: enter without a mouse, enlarge text, use a screen reader, activate captions, distinguish links and submit work from a modest device. A supplier's compliance statement is insufficient. The student who finds a barrier should not become the technician or wait for weeks. An equivalent, discreet and penalty-free alternative must exist while the problem is corrected.
Continuity must also be rehearsed. What happens if the connection fails during an assessment, the supplier closes or a file no longer opens? Essential materials need export in usable formats. The school needs backups, contacts, response times and another submission channel. An annual exercise, even in one class, exposes expired permissions and procedures that looked clear on paper.
A succession plan and an exit plan#
A change of teacher should not restart the project. A new colleague needs a short route through purpose, agreed practices, risks, instructions and support. A session connected to one real task is better than a tour of every feature. Afterwards, the colleague needs accompaniment and time for questions. Training is not teaching buttons; it explains which educational decisions the tool supports and which remain human.
The exit is designed on entry. The school should know how work, marks and settings will be downloaded; who verifies the copy; when the supplier will erase data; and what students and families will be told. Proprietary formats that prevent reuse should be avoided. Reversibility protects the school's negotiating position and allows it to remove a solution that no longer serves its purpose without losing years of educational memory.
When a tool fails, the response cannot depend on heroism. Record the incident, impact and resolution; determine whether it is isolated or a pattern; and change the procedure. Repeated errors provide information about the system. Penalising a pupil for a documented outage or blaming a teacher for opaque settings only hides the learning the organisation needs.
What families can ask#
A family does not need to audit servers. It can ask, “What does my child learn better with this tool?”, “What alternative exists if it fails?”, “Whom do we contact?”, “What data does it use?” and “When will it be reviewed?” Answers should be specific and consistent across classes. If every teacher improvises a channel, timetable and policy, the problem is not the family's lack of skill but the absence of a school-wide agreement.
Describe facts before taking positions: “This week we had three platforms and two night-time notices” is more useful than “there is too much technology”. Explain the impact—the teenager could not organise independently or needed the adult's phone—and suggest a testable change. Pupils should participate: they know where they become stuck, which notice they ignore through overload and which feature genuinely enables independent work.
A school can review a small set of evidence each term: incidents, tasks blocked by access, support hours, use of alternatives, perceived usefulness and effects on teacher workload. Results should be disaggregated where possible to detect when an average improvement conceals exclusion. No metric replaces speaking with people, but conversations without records allow each new school year to repeat the same mistakes.
What remains after enthusiasm should be less spectacular and more valuable: procedures anyone can follow, teachers with allocated time, families who know where to seek help, pupils who do not lose opportunities because of a password, and decisions that can be reversed. Mature innovation is not measured by how many tools remain. It is the capacity to preserve what helps, correct what causes harm and remove what no longer makes sense.
