Moving a school online with Moodle
When campuses closed, the schools that kept running had one thing in common: a real LMS behind them.
A learning management system is not video calls. It is the system of record that holds courses, enrolment, assignments, examinations, evaluation and grades together, so that teaching continues whether or not anyone can reach the building. Video conferencing replaces the lecture; it replaces nothing else. Moodle is open-source, deployed at institutional scale worldwide, and configurable down to the level of question types, grading scales and role permissions — which matters because no two institutions run exams the same way. The work of “moving a school online” is therefore mostly configuration and migration rather than software development: mapping your actual class structure, exam patterns, grading rules and staff roles onto a platform that can express them, then getting the existing material in without losing a term of history. Budget for that migration explicitly: converting a decade of question papers into structured question banks is teacher time rather than developer time, and it is the step that decides whether anyone logs in.
Why Moodle rather than a commercial platform?
Three reasons hold up once the novelty wears off. It bends to how you already teach: commercial platforms encode a single opinion about what a course is, and institutions with unusual exam patterns, multi-section subjects or internal-plus-external assessment tend to discover the mismatch only after the contract is signed, whereas Moodle’s activity and grading model is general enough to express most of what a real syllabus actually does. You own the data, too — grades, submissions and student records live in your own database, exportable in whatever format you choose, which for any institution carrying statutory record-keeping obligations is not a preference but a requirement. And there is no per-seat cliff: licence cost does not scale with enrolment, so growth becomes a hosting question rather than a budget renegotiation, and what you pay for instead is implementation and maintenance, a real cost but one you actually control rather than one set by a vendor’s pricing tier.
The honest counterpoint: Moodle out of the box is not a polished product. Left unconfigured it is genuinely unpleasant to use, and most complaints about it are complaints about an unconfigured install rather than about the platform.
What does “customised” actually mean here?
Customised means five specific, decisive things, not a vague promise of a nicer interface. Course management mapped to your actual classes: categories mirroring your departments and years, courses matching real subjects, and sections matching how the syllabus is actually divided, rather than a generic hierarchy teachers have to translate in their heads. Online exams built around question banks and timing: reusable banks organised by topic and difficulty, randomised selection and answer shuffling so no two attempts look identical, per-quiz time limits, attempt caps and defined review windows. Mixed grading, because almost every real exam is a mix: objective questions auto-graded on submission, long-form answers routed to the right teacher for manual marking with rubrics attached. A gradebook that publishes on your terms, with weighted categories reflecting the actual marking scheme, calculated aggregates, and controlled release so results appear when the institution intends. And roles that hold: separate, tested permissions for students, teachers, examiners and administrators, the part that gets skipped and then becomes an incident.
How long does an implementation take?
The build itself is rarely the constraint. A configured Moodle with courses, roles, exam settings and a working gradebook is a matter of weeks, not months; what stretches the timeline is content migration and training, both of which deserve their own budget line rather than being treated as an afterthought to the build. Migration is the honest unknown: existing material tends to live scattered across a decade of folders, personal drives and printed masters, and converting a stack of question papers into a structured question bank is teacher time rather than developer time, which makes it hard to estimate precisely and easy to underbudget. Training matters more than it sounds, because the staff who most need the system are usually the least interested in learning it, and adoption gets decided in the first fortnight rather than gradually. We plan for a pilot cohort — one department, one full assessment cycle — before rolling out institution-wide, because a failed exam day is not something a support article fixes.
What goes wrong most often?
Four things go wrong most often, roughly in order of frequency. Nobody defines the grading scheme before configuring the gradebook, so weightings get set by whoever happens to be clicking, and the numbers turn out wrong at the end of term when it matters most. Exam day is not load-tested: four hundred students submitting within the same ten minutes is a very different load profile from four hundred students browsing, and the only way to know a server will hold is to test the peak you will actually get. Roles get copied from a template rather than designed for the institution, and a teaching assistant turns out able to edit grades that have already been published. And backups are configured once and never restored, so an untested backup is really just a belief rather than a backup — restore one to a staging instance before you actually need it.
What does it look like when it works?
We built exactly this for a school during the COVID-19 lockdown, when students could no longer come to campus but academics, examinations and administration still had to keep running on schedule. The customised Moodle LMS carried the whole academic cycle in one place: course management, assignment submission, timed online exams, evaluation and grade publishing, configured to match the school’s actual departments, subjects and marking scheme rather than a generic template. The result was concrete rather than aspirational: 100% of academics moved online, with zero campus visits required at any point and 24/7 access to course material on a single platform, from a teacher setting up a lesson through to a student checking a published grade. Nothing about the underlying teaching changed; only the building it happened inside did, and the platform never became the bottleneck lockdown could easily have made it. The Lords Education case study has the full breakdown of how it was built and what it measured.
The pattern generalises past emergencies. Once assessment and records live in one system, the institution gets things it never had on paper: a real view of who is falling behind before the exam rather than after, and a grading process that survives a teacher leaving mid-term.
Where to go next
If you are weighing this up, our Moodle LMS development service covers implementation, migration and support. If the requirement is closer to a bespoke product than a configured platform — a student-facing app, video delivery, offline access — that is web applications and mobile app development territory, and the Shikshak EduTech case study shows what that path looks like across web, Android and iOS.