Generic safeguarding e-learning vs school-specific training

Generic safeguarding e-learning gets a reputation it does not entirely deserve. A well-made course covers the categories of abuse, the indicators an adult might notice, the duty to pass a concern on rather than investigate it, and the basic conduct rules about contact, communication and being alone with a child. Those foundations are shared across jurisdictions, and a school gains little by writing them again from scratch.

The format has practical strengths as well. A course sits on a server and runs at any hour, which matters when a new hire arrives in the second week of term and cannot wait for the next in-person session. The provider tracks changes to statutory guidance, so when the Department for Education publishes its September edition of Keeping children safe in education the course is updated without the school doing the reading. Per head, it is inexpensive. For a school with no DSL capacity and no policy worth building from yet, it is a reasonable starting position rather than a failure.

An outside voice has value too. Staff sometimes hear a difficult message about adult conduct more clearly from a source that has no history with their line manager. A generic module about grooming behaviours or about the boundary between affection and inappropriate contact can say things that are harder to say internally in a small staff room.

The gap opens at the point of action. Every generic course ends at some version of the same instruction: report your concern to the designated person without delay. It cannot name that person, the form they use, the deadline the school has set, or what happens when the person is on a flight to a conference. Staff finish the course knowing the principle and not the procedure, and the principle is the part they already had.

For an international school the gap is wider than that. The course was written for one legal system, and the host country has its own reporting duties, its own agencies and its own thresholds for involving them. The International Task Force on Child Protection was formed in 2014 precisely because international schools sit between systems and needed coordinated expectations, and the resources on the ICMEC Education Portal reflect that mixed picture. A module that names an English local authority designated officer teaches a route that does not exist in Astana, Dubai or Bogota.

Thresholds are the quietest failure. A generic course teaches that low-level concerns about adults should be recorded, which is correct, and says nothing about where this school draws the line between a note and an allegation. Two schools with identical policies on paper often behave differently here, and the difference is exactly what a new member of staff needs to learn. This is the argument for contextual safeguarding training at more length.

Then there is who the course reaches. Bus drivers, security officers, cleaners, kitchen staff and pool attendants see children in places teachers rarely go, and a course written for classroom teachers in English speaks to almost none of their working life. Schools that buy a licence for teaching staff and hold a briefing for everyone else have made a decision about risk without describing it that way, which is why training for non-teaching staff deserves its own design rather than a shortened version of the teacher module.

A school signs for a course in English because the leadership team works in English, and the assumption travels down the org chart until it meets a driver who reads Russian, a cleaner who reads Kazakh, and a teaching assistant whose Arabic is far stronger than her English. Those staff complete the module. Completion is what the system records, and comprehension is what the school needed.

Machine translation of a generic module produces a translated generic module, with the added risk that safeguarding vocabulary does not map cleanly between languages. Terms for guardianship, custody, consent, neglect and the local reporting authority carry legal meaning that a general translation engine flattens. A school that translates the module without translating and checking the underlying policy has produced a document nobody can act on with confidence.

The practical test is simple to run and uncomfortable to fail. Ask five staff who do not work in English, in their own language, who they tell if a child says something worrying, and by when. If the answers vary, the training language is a control failure rather than a convenience issue. The same reasoning applies to the policy itself, and I have written separately about translation compared with localisation.

Quotes for generic e-learning look cheap because the visible number is a per-seat licence. The full cost has three more lines. The first is administrative time chasing completions across a staff body with high turnover. The second is the DSL time spent writing the local supplement that the course does not cover, which is real work that rarely appears in any budget. The third is renewal, which scales with headcount every year and does not fall as the school gets better at safeguarding.

School-specific training inverts the shape. The first year carries an up-front block of leadership and DSL time to get the policy into a state worth training from, plus authoring or translation in each working language, plus whatever the platform or author costs. The second year is cheaper, because the school edits versions rather than writing from scratch, and the third year is cheaper again once the policy has stabilised. Comparing a single year favours the licence; comparing three years usually does not.

Staff time dominates both routes, and it is the line schools most often leave out. A school of 120 staff spending 45 minutes each on annual training spends 90 hours of paid working time, before anyone reviews an answer. That is the reason a shorter module built from the school's own procedure can be a better purchase than a longer generic one: the same 90 hours buys practice of the actual reporting route instead of a second pass through the categories of abuse. Put DSL hours in the comparison at a real internal rate, and count them on both sides.

I am not going to offer a sector-wide figure for either route, because I do not have one I can defend. The arithmetic above is the method: licence plus admin plus supplement plus renewal on one side, authoring plus translation plus platform plus maintenance on the other, staff hours on both, over three years rather than one.

The arrangement that works in practice puts the two routes in sequence rather than in competition. A generic course carries the shared foundations and runs in the first fortnight of induction, so a new arrival has the concepts before their first week with children ends. A short school-specific module follows within the first month and covers only what the generic course could not: the named DSL and deputies, the form, the deadline, the local agencies, the thresholds this school uses, and the two or three situations that are specific to the campus, such as the boarding house at eleven at night or the bus route home.

Keep the school layer short. Twenty minutes of material that staff repeat every year, with the scenarios rotated, holds attention better than an hour they resent. Put the assessment on the school layer rather than on the foundations, because the foundations are where multiple choice is least revealing and the procedure is where a wrong answer has consequences. Certificates should name both parts, so a receiving school can see which provider covered the concepts and which policy version covered the practice.

Review the split annually and expect it to move. As the school's own policy improves, more of the generic course becomes redundant, and some schools eventually keep the licence only for induction and for specialist topics they do not want to author, such as online harms. That drift is a sign the safeguarding work is going well rather than a procurement error.

Whichever route a school takes, the same seven questions separate a working arrangement from an expensive one. Ask them of a vendor and of your own leadership team, and write down the answers, because the gaps are usually visible before anything is signed.

• How many of my staff cannot read this course in the language it ships in, and what happens to those people?
• Where in the course is the name of the person my staff must tell, and the deadline for telling them?
• What happens to the content when my child protection policy is republished in June?
• Can I see what a member of staff answered, or only that they reached the end?
• Does the certificate name the policy version the training was built from?
• What does the third year cost at my projected headcount, including administrative and DSL hours?
• Who approved this content for my school, by name, and on what date?

I have worked in international education since 2008, most recently as CIS accreditation coordinator at Miras International School in Astana, and the platform started because that school's training was standard modules written for another country while its staff worked in three languages. SafeguardIS builds training from the school's own safeguarding and child protection policies, so what staff learn is what that school actually does. The Designated Safeguarding Lead reviews and approves every module before staff see it, publishing requires a named reviewer's declaration, and the approved content is hashed and logged.

Training ships in English, Russian and Kazakh today. If your staff work in a language the platform does not ship yet, I build that language in for your school as part of the pilot. Certificates are countersigned by the Head of School and the DSL and carry a code anyone can check on a public verification page. Staff can ask the AI policy assistant questions that are answered from the school's own policies, while the DSL keeps every safeguarding decision and no child's details belong in a chat window. Miras International School is the pilot school.

A policy audit feature checks a school's documents against ITFCP-aligned expectations and produces an action plan, which is the sensible first step when a school suspects its policy is not yet strong enough to train from. I also run live workshops in English and Russian and whole-school safeguarding audits. If you are weighing a licence renewal against building your own layer, or working through how to choose safeguarding training software, book a 20-minute walkthrough and bring your current policy.