Plenty of strong school robotics clubs are led by teachers who cannot code. The person running the club is often a geography, art or life-orientation teacher who volunteered, got handed a box of kits, and worked it out as they went. You do not need a computer science degree to do this well. You need a way to stay slightly ahead of your learners, a plan that puts them at the centre of the work, and a short list of people to call when something breaks.

TL;DR

• Facilitate, do not lecture. Your job is to keep learners unstuck and moving, not to know every answer in advance.
• Stay one lesson ahead. Do each build yourself the week before, and note what went wrong.
• Recruit student technicians. Two or three learners can handle charging, cables and first-line debugging.
• Run projects, not talks. Set a challenge with a simple rubric and let learners find their own path.
• Never be the only source of help. Lean on vendor training, holiday workshops and online communities.

Change the job description first

The biggest shift is mental. A robotics club does not need a lecturer at the front explaining code. It needs a facilitator who keeps the room productive. Think of a sports coach who was never a professional player. They set drills, watch closely, ask good questions, and step in when someone is stuck. They do not have to be the best striker on the field to develop one.

Once you accept that you are the coach and not the star, the pressure drops. You are allowed to say you do not know and suggest finding out together. Learners copy a teacher who models calm, curious debugging, and they freeze around one who pretends to know everything.

Stay one lesson ahead

This is the whole trick, and it is boring on purpose. You do not need to master the subject. You need to be one week in front of the class. Each week, sit down for thirty to forty minutes and do the exact build your learners will attempt. Type the code, run it, break it, fix it. Keep three things as you go:

• A screenshot or photo of the working result, so you know what finished looks like.
• A short list of things that went wrong. The mistakes you make are almost exactly the ones your learners will make.
• Two or three checkpoint questions you can ask to nudge a stuck group without handing over the answer.

Block-based coding makes this realistic for a non-specialist. Dragging blocks together removes the typos and missing punctuation that eat time in a text language, so you can focus on the logic and the build rather than on syntax.

Recruit student technicians

You cannot be at ten tables at once, so do not try. Every club has a few learners who race ahead and finish early. Give them a job. Appoint two or three student technicians on a rotating basis, and make it a position of status rather than a punishment for being fast.

Technicians handle the tasks that would otherwise swallow your lesson: charging kits before the session, sorting cables, checking batteries, and being first responders when a group's robot will not move. An ask three, then me rule works well. Learners ask a neighbour, then a technician, then you. By the time a problem reaches you it is usually a real one worth your attention, and half the room has practised debugging on the way. This is not offloading work you should be doing; explaining a fix to a peer is one of the strongest ways to learn, so your technicians gain the most of anyone in the room.

Run projects, not lectures

A term of step-by-step worksheets teaches copying. A term of projects teaches problem solving, and it quietly hides the fact that you are not an expert, because there is no single script for you to get wrong. Give a goal and a constraint, then get out of the way.

Good beginner challenges are physical and easy to judge by eye: a robot that follows a taped line, a burglar alarm that reacts to movement, a device that shows the temperature and warns when a room gets too hot. Learners can tell for themselves whether it works, which means you do not have to be the arbiter of correctness.

A simple rubric beats a right answer

Because there is no single correct solution, mark the process instead of a model answer. A rubric any teacher can apply might score four things equally: does it work, is the code tidy and labelled, did the group test and improve it, and can they explain how it works. That last column matters most. A group that can explain a modest line-follower understands more than a group with a flashy robot they copied and cannot describe.

Choosing kit you can actually support

Support is easier when you standardise. Pick one board type for the whole club rather than a shelf of mixed hardware, so every problem you solve applies to every table. Favour kits with strong block coding, clear guides and parts you can reorder locally. A beginner-friendly board such as the sheenbot∞ keeps a first term about sensors, motors and lights rather than syntax. Budget roughly 10 to 15 percent of the kit cost for spare cables, batteries and a few replacement parts, because cables fail first and always at the worst time.

How you deploy the kit matters as much as which kit you buy. Three common models:

Model	What you need	Good for	Watch out for
Shared class set, learners in pairs	Fewer kits, one board type, a storage box	Tight budgets and a first year	One learner can dominate; assign a driver and a navigator and swap them
One kit per learner	More kits, more charging and storage	Smaller or older, confident groups	Higher spend and more hardware to maintain each week
Screen first, hardware later	Only the computer lab you already have	Testing interest before you commit budget	Learners want the physical build soon, so plan the switch early

If you are not sure the interest will hold, a shared class set of about ten kits with learners in pairs is the safest way to start, and you can grow into one-to-one later. When you are ready to buy, a supplier that also offers curriculum and support is worth more than the cheapest box; you can see what a class set looks like in the store.

Never be the only source of help

The fastest way to burn out is to be the single point of knowledge for the whole club, so build a support network before you need it. Vendor training is the most direct route. A good supplier runs sessions for teachers, not just learners, and can get a nervous first-timer confident in an afternoon. Booking a free trial lesson lets you watch how a session is actually facilitated before you build your own, and the winter holiday workshops in July are a good place to send keen learners or to sit in and learn the ropes yourself. Online communities of teachers and makers are the other pillar; nearly every wiring problem or block-logic puzzle you meet has already been solved and written up by someone else.

A workable session shape

Keep every session on the same rhythm so learners settle quickly and you spend less energy managing the room. A forty-minute club session can run as five minutes to recap and set the challenge, thirty minutes of hands-on building while you and the technicians circulate, and five minutes for two groups to show progress and name one thing they will try next week. Charge the kits the day before, because relying on a top-up during the meeting is a gamble when load shedding lands on club afternoon.

Takeaway

You do not need to be a computer teacher to run a robotics club that learners love. You need to stop performing as the expert and start working as the coach. Stay a week ahead, hand real responsibility to student technicians, build the term around projects with a simple rubric, and keep a support network within reach. The coding and robotics curriculum reaching South African schools was never meant only for specialists, and the clubs that thrive are usually the ones where the adult in the room is learning too.

FAQ

Do I need to know how to code before I start?

No. You need to be willing to try the weekly build before your learners do, and to stay a lesson ahead. Block-based coding removes most of the typing errors, so you can learn alongside the club rather than years in advance of it.

What if a learner knows more than I do?

Treat it as a gift. Make that learner a technician or a mentor for a table. Your value is not superior knowledge; it is structure, fairness, pacing and keeping the whole room moving. A club runs on organisation more than expertise.

How many kits do I need to begin?

A shared set of about ten kits with learners working in pairs suits most starting clubs. Standardise on one board type so every fix you learn is reusable, and keep a small stock of spare cables and batteries so one failure does not stop the session.

How long should a club session be?

About forty minutes to an hour works well. Long enough to build and test something, short enough to stay focused after a full school day. A predictable structure of recap, build and share helps more than extra time on the clock.

Where can I get help when something breaks?

Start with your supplier's training and guides, then online teacher and maker communities. Booking vendor training or a workshop early in the year pays off, because you meet the tricky problems in a calm setting instead of in front of a stuck class.