Screens down, circuits up: choosing a first robotics kit

I've run a weekend STEM club in Lekki for six years, and I can tell you the most common fate of a first robotics kit: built once, on Boxing Day, by an enthusiastic uncle, while the child watched. By February it's under the bed. The kit wasn't bad — it was wrong. Wrong for the child's age, wrong for their patience, and usually chosen by its box, which showed a robot far more capable than anything inside.

The fix is to stop shopping for a robot and start shopping for a first success. The question isn't 'what can this kit do?' — it's 'what can my child do with it, alone, in the first twenty minutes?' Because that first unassisted win is what decides whether the kit becomes a hobby or landfill.

Match the kit to the child, not the box

For ages five to seven, skip code entirely: you want screen-free logic robots that are programmed by pressing arrow buttons or laying physical tiles. The child learns sequencing — the real foundation of programming — without a tablet in sight. For eight to ten, look for snap-fit builds and block-based coding like Scratch: motors, sensors and drag-and-drop logic, with results visible in minutes rather than evenings.

From eleven up, choose a kit that grows: one that starts in block coding but opens up to Python or Arduino when the child is ready. This is the age where a ceiling matters more than a floor — a kit they can't outgrow in a term is worth two they can. And at every age, check the battery situation before you buy. I have watched more robotics dreams die from missing AA batteries than from any lack of talent.

The best first robot is not the smartest one on the shelf — it's the one your child can make do something, alone, in the first twenty minutes.

The red flags

Avoid kits that are apps in disguise — where the 'robot' is really a Bluetooth speaker on wheels and all the thinking happens on a screen your child watches. Avoid single-build kits that can only ever become the one robot on the box; rebuildability is where the value lives. And be wary of kits demanding an adult for every step. Some parent involvement is lovely; total dependence means the box has bought you a chore, not your child a hobby.

Also check the sensor list, not the sticker count. One distance sensor a child truly understands beats six components that remain mysterious. Mystery is fine in a magic set; in an engineering kit it breeds helplessness.

Your job after the box is open

Be the assistant, never the lead engineer. Read the manual aloud if asked, hold the piece they can't reach, and be impressed on demand — but let their hands do the building and their logic make the mistakes. A robot that turns left when it should have turned right is not a failure; it is the single best debugging lesson available at any price.

And when the guided builds run out, set challenges instead of buying the next kit straight away: make it follow a line of black tape across the parlour floor, make it stop before hitting the wall, make it dance. A child who has bent a kit past its instructions is no longer following STEM — they're doing it.