A school does not need to wait for a crisis to know that its technological change is going wrong. The signal we should not ignore is a recognisable combination: students understand less or participate less well, teachers spend more time operating tools than teaching, and some families begin to be excluded. No single indicator proves failure, but when these three effects coincide, the school should pause expansion, listen and review the design.
Technology is a means, not the objective of the course#
Imagine a secondary school that replaces notebooks, notices and part of assessment with a platform. During the first weeks everything looks modern: more submissions are recorded, charts appear instantly and messages are automated. Yet several students say in tutorial time that they no longer know what matters most; one teacher spends every afternoon resolving access problems; two families receive notices after the deadline. Leaders see that the number of clicks is rising and conclude that adoption is succeeding. That is the signal: digital activity is being measured, but nobody is checking whether learning, relationships or autonomy have improved.
Imagine a secondary school that replaces notebooks, notices and part of assessment with a platform.
The professional question is not whether a school uses too much or too little technology, but which educational problem it intended to solve. If the purpose was faster feedback, measure the time and usefulness of comments, not logins. If it was participation, examine who contributes, who withdraws and why. If it was less bureaucracy, count the total time of teachers, administration, students and families. A tool can work technically and fail pedagogically. Defining an observable outcome beforehand allows the school to recognise that difference without turning discussion into a battle between supporters and opponents of screens.
First signal: understanding falls while submissions rise#
An early warning occurs when students complete tasks but explain their learning less well. They may copy instructions, follow steps or answer quizzes without constructing an idea of their own. A teacher can test this by asking for an oral explanation, a new example or a reasoned correction of an error. The device is not automatically the cause: perhaps the activity is too fragmented, reading is uncomfortable or automatic feedback rewards quick attempts. To investigate, compare the same skill in different formats for several weeks and review actual student work, rather than relying only on aggregate grades.
Families may notice something similar at home: a young person says the work is finished but cannot summarise it; switches constantly between windows; needs continuous instructions; or becomes anxious when the application changes. The answer should not be to monitor every movement or remove all technology. It is more useful to ask, “What were you meant to learn?”, “Which part did you understand?” and “Where did you get stuck?” These questions distinguish difficulties with content, organisation, accessibility or self-regulation. Taking two or three concrete examples to a tutorial is more useful than declaring that screens distract or the platform is useless.
Second signal: invisible work displaces teaching#
The change deserves review when teachers spend increasing time on passwords, incompatible formats, incidents, duplicated marks or manual account creation. This work remains hidden because it is distributed across minutes at the start of every lesson and messages outside working hours. A school can measure it with a short two-week log: type of incident, people affected, minutes lost and remedy used. The aim is not to monitor teachers, but to discover whether the system transfers costs to the people sustaining the classroom. Ten lost minutes per group each day can consume a substantial share of learning time.
Teachers' cognitive load also matters. Introducing five tools at once means remembering permissions, workflows and exceptions while attending to a diverse group. Errors increase even among skilled professionals. A healthy implementation reduces variables: one purpose for each tool, common instructions, identifiable support and a fallback that lets the lesson continue. If everything depends on one enthusiastic person answering every question, the school does not have a system; it has a heroic effort that will be difficult to maintain.
Third signal: the same people are always left outside#
Technological inequality is not limited to lacking a device. It includes unstable connectivity, a shared phone, little space, work schedules, disability, language, low digital confidence or family circumstances that make notices difficult to follow. The serious signal is that the same people submit late, miss communications or repeatedly have to justify their reality. If a fallback exists but requires asking publicly for a favour, printing far from the classroom or losing marks, it is not equivalent. Inclusion is tested by walking through the complete task from the perspective of the person encountering the barrier.
A school can conduct a simple audit without labelling anyone. Choose three processes—receiving a notice, submitting a task and requesting help—and test them with a basic phone, keyboard, screen reader, interrupted connection and locked account. Then ask students and families which steps made them hesitate or give up. Responses should be confidential and lead to visible changes. Asking questions without reporting results erodes trust; explaining “this is what you told us and this is what we changed” turns participation into genuine school governance.
Fourth signal: data outrank educational judgement#
A platform may display connected time, progress, risk or behaviour. These data are partial signals, not diagnoses. A student may leave a window open, work offline, share a device or need more time because of a barrier. Trouble begins when a number triggers punishment, a label or family contact without human review. Every automated alert should state what it measures, what it cannot know, who checks it and how the affected person can provide context. Decisions affecting assessment, behaviour or support should never rest on an opaque score.
Privacy offers signals too. If nobody can explain which data are collected, how long they remain, who has access or how they are deleted, a basic condition of accountability is missing. Schools should not ask children to accept terms as though they were ordinary consumers. Before using a service, the institution must assess necessity and proportionality, choose the least intrusive settings and provide understandable information. The Spanish Data Protection Agency's guide for schools helps organise responsibilities and common situations.
What to do when the signal appears#
Pausing expansion does not mean turning back or assigning blame. It means opening a bounded review. First, describe the problem with facts: which task, which group, since when and with what consequences. Next, protect the immediate activity through an alternative route. Then listen separately to students, families and professionals, because voices with less power often stay silent in a large meeting. Finally, choose whether to adjust, train, limit, replace or withdraw, and set a date for checking whether the measure worked.
A useful review uses a small, balanced set of indicators. For learning: a product that demonstrates understanding, not merely submission rate. For equity: who cannot complete the journey and which barrier they encounter. For wellbeing: interruptions, frustration or pressure outside expected hours. For work: minutes of support and duplication. For protection: incidents, permissions and deletion requests. Numbers should be accompanied by short accounts, because an average may hide a small group experiencing intense harm.
Agree thresholds before the pilot. For example, if access problems waste time in more than a specified number of sessions, if the alternative does not reach the same objective, or if academic improvement does not appear within the defined period, the tool is not expanded. The threshold need not be universal; it must be explicit and reviewable. This prevents a poor decision from continuing because of pride, money already spent or fear of appearing insufficiently innovative.
The role of each community member#
Students are not passive sensors. They can help with design by explaining which instructions they understand, which notifications help and when a tool interrupts concentration. To speak freely, they need assurance that criticising a platform will not affect their marks. A small, diverse group can test tasks and document barriers, but should not carry responsibility for solving the school's technical problems. Their experience informs the decision; responsibility remains adult and institutional.
Families need a clear channel and a concrete request. Instead of expecting them to master every application, the school can provide one short page: purpose, access, data used, alternative, support contact and expected hours. It can also ask whether notices arrive, whether the device is shared and how much support the task requires at home. A family should not become a technical helpdesk or permanent supervisor. If an activity only works with continuous adult oversight, its design should be reviewed.
Teachers need protected time to learn, agree criteria and examine evidence. Training based only on buttons quickly expires. It is more sustainable to practise with a real activity, anticipate errors, design an accessible option and decide which evidence will demonstrate learning. Teacher digital competence includes pedagogical and ethical judgement, not merely operational skill. The INTEF framework for teachers' digital competence provides an official reference for planning that professional development.
A thirty-day protocol#
In the first week, the team defines the educational objective and records a baseline: a sample of work, time spent, incidents and people who cannot complete the process. In the second, it tests the change with a limited group and preserves an equivalent non-digital route. In the third, it collects evidence and holds short conversations with students, families and teachers. In the fourth, it compares results with the starting point and documents a decision. Timing can vary, but the cycle should be short enough to correct and long enough to move beyond normal early difficulties.
The final decision fits on one page: purpose, participants, observed benefit, harms or burdens, affected groups, data processed, total cost, adjustments, owner and next review. It should also state what happens if the supplier changes conditions or connectivity disappears. UNESCO's report on technology in education proposes judging technology through relevance, equity, scalability and sustainability; this framework prevents novelty from being confused with progress.
A practical traffic light for deciding#
Green appears when a defined need improves, the fallback works, teachers can sustain the practice, students understand more and barriers diminish. Amber appears when benefits exist but costs are concentrated: scope is limited and problems are corrected before expansion. Red appears when understanding falls, one group is repeatedly excluded, data drive opaque decisions, there is no exit, or support work consumes teaching. The traffic light does not replace judgement; it requires the evidence behind that judgement to be named.
The European Commission's Digital Education Action Plan 2021–2027 places quality, inclusion and accessibility at the centre of transformation. In school practice, that requires something simple and demanding: the tool must adapt to the purpose and the people, not the reverse. A mature community can retain useful technology, modify it or leave it without turning the choice into a question of identity.
The signal we should not ignore is neither an isolated fault nor the normal resistance involved in learning something new. It is a persistent pattern: less understanding, more invisible workload and repeated exclusion. Detecting it early protects students, families and teachers. A responsible response combines evidence, listening, an alternative and a review date. Technological change deserves to be called educational only when it expands the capacity to learn, relate and participate among the people who make up the school.
